AI data analysis method and system based on bank industry customer marketing

Abstract

The invention discloses an AI data analysis method and system based on bank industry customer marketing, and the method comprises the steps: collecting customer information from a plurality of data sources, obtaining original data, and carrying out the cleaning and integration of the original data; through performing weight score assignment and dynamic adjustment on customer multi-dimensional information, a marketing value score of a customer and a customer and product matching degree score are calculated in combination with an algorithm model, and accurate marketing positioning is realized in an auxiliary manner. Generating an initial marketing strategy based on a calculation result, supporting strategy adjustment and optimization through a multi-round dialogue mechanism, and determining an optimal marketing scheme; the successful marketing cases are stored through the expert knowledge base, and reference suggestions are provided for new marketing decisions. The data analysis and decision-making ability in the customer marketing process is enhanced by using the artificial intelligence technology, and the problems of inaccurate customer identification, unreasonable marketing resource allocation and weak adaptability to the complex market environment in the traditional marketing method are solved.

Description

Technical Field

[0001] This invention relates to the field of bank customer marketing, and in particular to an AI data analysis method and system based on bank customer marketing. Background Technology

[0002] Traditional bank customer marketing relies mainly on the experience and judgment of sales personnel and traditional analytical tools, adopting a batch marketing approach. This approach suffers from problems such as weak targeting, low efficiency, insufficient accuracy, and poor dynamic adjustment capabilities, making it difficult to adapt to the rapidly changing market environment.

[0003] Traditional bank customer marketing relies mainly on the experience and judgment of sales personnel and traditional analytical tools, and adopts a batch marketing approach.

[0004] Existing technologies suffer from problems such as weak targeting, low efficiency, insufficient accuracy, and poor dynamic adjustment capabilities, making it difficult to adapt to the rapidly changing market environment. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide an AI data analysis method and system for banking customer marketing that overcomes or at least partially solves the above problems.

[0006] According to one aspect of the present invention, an AI data analysis method based on banking customer marketing is provided, the data analysis method comprising: Customer information is collected from multiple data sources to obtain raw data, which is then cleaned and integrated. By assigning weights and dynamically adjusting multi-dimensional customer information, and combining algorithm models to calculate the customer's marketing value score and customer-product matching score, we can assist in achieving precise marketing positioning. An initial marketing strategy is generated based on the calculation results, and the strategy can be adjusted and optimized through a multi-round dialogue mechanism to determine the optimal marketing plan. Successful marketing case studies are stored in an expert knowledge base, providing reference suggestions for new marketing decisions.

[0007] Optionally, the customer information includes basic customer information, transaction information, business registration information, public opinion information, and risk information.

[0008] Optionally, the step of collecting customer information from multiple data sources to obtain raw data, and cleaning and integrating the raw data specifically includes: Customer information is obtained from multiple data sources, including internal business systems, publicly available external data, and data from third-party service providers; The collected data is cleaned and preprocessed, including missing value imputation, outlier handling, and data standardization.

[0009] Optionally, obtaining customer information from multiple data sources, including internal business systems, publicly available external data, and data from third-party service providers, specifically includes: Internal customer data acquisition: Obtain customer basic information tables, account information tables, and transaction records from the enterprise's core business systems; extract data daily using ETL tools, imput missing values ​​using the mean imputation method, and identify and correct outliers using the 3σ criterion; after data cleaning, load the data into the customer subject model of the data warehouse. External publicly available data processing: Obtain business registration information data daily via API interface, including company registered capital, business scope, and shareholder information; We collect public opinion information hourly using web crawlers, including customer-related public opinion data from news, social media, and forum channels; all external data is cleaned, transformed, and then stored in the corresponding data model. Third-party service provider information processing: Receives customer credit ratings and credit limits from third-party institutions daily through the data exchange platform.

[0010] Optionally, the step of weighting and dynamically adjusting multi-dimensional customer information, and calculating the customer's marketing value score and customer-product matching score using an algorithm model, specifically includes: Calculate customer value score: Among them, D i W represents the score of the i-th information item. i Indicates the corresponding weight; Customer value calculation: The system automatically performs customer value calculation tasks every day at midnight. First, it obtains the latest customer information from various data models, then calculates the score for each customer according to the preset scoring rules, and finally generates a customer value analysis report according to industry and region dimensions. Product matching degree calculation: When an account manager initiates a product recommendation request, the matching degree between the specified customer and the available products is calculated in real time. During the calculation process, factors such as the customer's historical transaction preferences, risk tolerance, and investment period requirements are taken into account, and a matching degree score and recommendation reasons are given in combination.

[0011] Optionally, the step of generating an initial marketing strategy based on the calculation results, and supporting strategy adjustment and optimization through a multi-round dialogue mechanism to determine the optimal marketing plan, specifically includes: Multi-round dialogue decision-making process: Account managers interact with the system through a web interface; The initial marketing strategy proposals were presented, and the account manager requested adjustments. By understanding the requirements through natural language processing, adjusting the strategy parameters in real time, and providing a new strategy solution within 5 seconds; Strategy Effectiveness Evaluation: Offers strategy simulation functionality, allowing account managers to input different market scenarios and predict the expected performance of strategies under various market conditions based on historical data and machine learning models. Optionally, the expected results include expected rate of return, customer response rate, and risk indicators.

[0012] Optionally, storing successful marketing cases in an expert knowledge base and providing reference suggestions for new marketing decisions specifically includes: Automatic case archiving: The system automatically collects campaign performance data after the marketing campaign ends; If the activity achieves the preset excellent standard, the case will be automatically archived in the expert knowledge base; Knowledge Base Intelligent Recommendation: When account managers develop new marketing strategies, the system automatically retrieves similar success stories from the expert knowledge base and recommends relevant strategy parameters and precautions. It supports searches based on multiple features, including customer type, product category, market environment, and risk preference.

[0013] Optionally, the activity performance data includes actual return on investment, customer response, and complaint rate indicators.

[0014] This invention also provides an AI data analysis system based on banking customer marketing, applying the aforementioned AI data analysis method for banking customer marketing. The data analysis system includes: The data acquisition module is used to collect customer information from multiple data sources, obtain raw data, and clean and integrate the raw data. The intelligent analysis module is used to assign weights and dynamically adjust customer information across multiple dimensions, and combine algorithm models to calculate the customer's marketing value score and customer-product matching score, thereby assisting in achieving precise marketing positioning. The AI ​​decision-making module is used to generate initial marketing strategies based on calculation results, and supports strategy adjustment and optimization through multi-round dialogue mechanisms to determine the optimal marketing plan; The expert knowledge base module is used to store successful marketing cases and provide reference suggestions for new marketing decisions.

[0015] This invention provides an AI data analysis method and system for customer marketing in the banking industry. The data analysis method includes: collecting customer information from multiple data sources to obtain raw data, and cleaning and integrating the raw data; assigning weights and dynamically adjusting multi-dimensional customer information, and calculating the customer's marketing value score and customer-product matching score using an algorithm model to assist in achieving precise marketing positioning; generating an initial marketing strategy based on the calculation results, and supporting strategy adjustment and optimization through a multi-round dialogue mechanism to determine the optimal marketing plan; storing successful marketing cases through an expert knowledge base and providing reference suggestions for new marketing decisions. This invention utilizes artificial intelligence technology to enhance data analysis and decision-making capabilities in the customer marketing process, solving problems such as inaccurate customer identification, unreasonable allocation of marketing resources, and weak adaptability to complex market environments in traditional marketing methods.

[0016] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0018] Figure 1 A flowchart illustrating an AI data analysis method for banking customer marketing provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the analysis method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of a multi-turn dialogue decision-making process provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the expert knowledge base update process provided in an embodiment of the present invention; Figure 5 This is a block diagram illustrating the composition of an AI data analysis system for banking customer marketing, provided as an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0020] The terms "comprising" and "having," and any variations thereof, in the specification, embodiments, claims, and drawings of this invention are intended to cover non-exclusive inclusion, such as including a series of steps or units.

[0021] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0022] Example 1 like Figure 1 As shown, an AI data analysis method and system for banking customer marketing is presented. The data analysis method includes: collecting customer information from multiple data sources to obtain raw data, and cleaning and integrating the raw data; assigning weights and dynamically adjusting multi-dimensional customer information, and calculating the customer's marketing value score and customer-product matching score using an algorithm model to assist in achieving precise marketing positioning; generating an initial marketing strategy based on the calculation results, and supporting strategy adjustment and optimization through a multi-round dialogue mechanism to determine the optimal marketing plan; storing successful marketing cases through an expert knowledge base and providing reference suggestions for new marketing decisions.

[0023] like Figure 2 As shown, the steps of the analysis method include: Step S201: Obtain marketing-related data, perform data collection, preprocessing, feature extraction, and multi-dimensional correlation analysis to generate market insight results; Step S202: Based on historical marketing data from the market insight results set, use statistical analysis methods to predict the potential effects of the marketing plan and generate marketing prediction results; Step S203: Based on market insights and marketing forecasts, use a multi-turn dialogue management mechanism to perform intelligent reasoning and optimization calculations on different marketing strategies; Step S204: Update the expert knowledge base based on the optimal decision-making solution to build a reusable marketing knowledge system.

[0024] like Figure 3 As shown, the multi-round dialogue decision-making process includes: Step S301: Decision initialization. Based on the preliminary results such as customer value and product matching degree output by the intelligent analysis module, the system generates an initial marketing strategy plan. Step S302: User input instructions. The user submits optimization and adjustment instructions for the initial solution through a natural language interface. Step S303: Instruction parsing and reasoning, the core of the AI ​​decision-making module—the large language model performs semantic parsing and intent recognition on user instructions, transforming them into operable structured parameters; Step S304: Strategy optimization calculation and response generation. Based on the reasoning results, the system calls the internal algorithm model to quickly recalculate and optimize, generating an updated marketing strategy plan. Step S305: Decision loop determination. The system determines whether the current solution meets the user's needs or whether the number of dialogue rounds has reached the preset limit. If not, steps S302 to S304 are executed repeatedly until the optimal decision solution approved by the user is output. Step S306: Output the final decision solution. Output the final marketing strategy solution after multiple rounds of dialogue and optimization to the user, and provide a detailed implementation proposal.

[0025] like Figure 4 As shown, the expert knowledge base update process includes: Step S401: Marketing strategy execution and performance tracking; Step S402: Effectiveness evaluation and case generation; Step S403: Feature extraction and labeling; Step S404: Include the cases in the database; Step S405: Knowledge retrieval and recommendation.

[0026] like Figure 5 As shown, the data acquisition module is used to collect customer information from multiple data sources, including basic customer information, transaction information, business registration information, public opinion information, and risk information, and to clean and integrate the raw data to support subsequent analysis and decision-making.

[0027] Intelligent Analysis Module: By assigning weights and dynamically adjusting multi-dimensional customer information, and combining algorithm models to calculate the customer's marketing value score and customer-product matching score, it assists in achieving precise marketing positioning.

[0028] AI Decision Module: Based on the results of data collection and intelligent analysis, it integrates the causal reasoning capabilities of structured data attribution and large language models, and uses a multi-turn dialogue mechanism to intelligently deduce and optimize different marketing strategies in order to determine the optimal marketing solution.

[0029] Expert knowledge base: In addition to the system's built-in marketing models, it supports users to import new knowledge models, expert experience, and marketing cases to continuously optimize analysis results and decision-making suggestions, making them more in line with actual business needs.

[0030] The marketing decision data analysis method based on artificial intelligence provided by this invention includes: acquiring marketing-related data, performing data collection, preprocessing, feature extraction and multi-dimensional correlation analysis, and generating market insight results; the data includes basic customer information, transaction information, business registration information, public opinion information, risk information and credit information; the market insight results include target customer preferences, customer risk and customer value. Customer value V is calculated using the following formula: Where D is the score corresponding to a certain customer information item, and W is the value weight of the information item. For example, the value of a customer is calculated based on its company size, customer type, and registered capital, with the following corresponding score weights: Customer A is a medium-sized bank with a registered capital of 70 billion. Then V = 100 * 10% + 100 * 10% + 60 * 18% = 30.8. Based on the market insights and historical marketing data, statistical analysis methods are used to predict the potential effects of the marketing plan and generate marketing forecast results, including the marketability of target customers, marketing value, and product preferences.

[0031] By combining customer type, company size, registered capital, and the customer's scale in the product, the matching degree between the customer and the product is calculated.

[0032] For example, if Customer A's product scale is 30 billion yuan, then its matching degree is calculated as: V = 100×10% + 100×10% + 60×18% + 60×30% = 48.8. The relevant information items for product matching degree calculation are as follows: Based on market insights and marketing forecasts, a multi-turn dialogue management mechanism is used to perform intelligent reasoning and optimization calculations on different marketing strategies. In a preferred embodiment, the system optimizes strategies through the following steps: (1) Initialize marketing strategy parameters, including target customer groups, marketing budget, and expected rate of return; (2) Receive adjustment instructions from users through multi-turn dialogues, including weight modification and scoring standard update; (3) Use a large language model to perform causal reasoning and evaluate the impact of different parameter adjustments on marketing effectiveness; (4) Output the optimized marketing strategy plan, including a recommended customer list, product mix, and budget allocation suggestions.

[0033] The expert knowledge base is updated based on the optimal decision-making solution to build a reusable marketing knowledge system. In practice, the system will perform the following operations: Record the key parameters and final results of this marketing decision; (2) When the marketing effect reaches the preset success standard, the decision-making plan will be automatically archived to the expert knowledge base; (3) Add feature tags to the cases, including market environment type, customer group characteristics, and product type; (4) Establish a case retrieval mechanism to support fast query and matching based on feature tags.

[0034] Example 2 Data acquisition and processing, including: Internal customer data acquisition: Customer basic information tables, account information tables, and transaction logs are obtained from the enterprise's core business systems. Data is extracted daily using ETL tools; missing values ​​are imputed using the mean, and outliers are identified and corrected using the 3σ criterion. After data cleaning, the data is loaded into the customer subject model in the data warehouse.

[0035] External publicly available data processing: Business registration information, including registered capital, business scope, and shareholder information, is acquired daily via API. Public opinion information, including customer-related data from news, social media, and forums, is collected hourly via web crawlers. All external data is cleaned, transformed, and then stored in the corresponding data models.

[0036] Third-party service provider information processing: The system receives customer credit ratings, credit limits, and other information daily from third-party institutions through a data exchange platform. After receiving the data, it undergoes format conversion, validation, and deduplication before being loaded into the risk rating model.

[0037] Intelligent analysis, including: Customer Value Calculation: The system automatically performs customer value calculation tasks every day at midnight. First, it retrieves the latest customer information from various data models, then calculates a score for each customer according to preset scoring rules, and finally generates a customer value analysis report based on dimensions such as industry and region.

[0038] Product matching degree calculation: When an account manager initiates a product recommendation request, the system calculates the matching degree between the specified customer and the available products in real time. The calculation process considers factors such as the customer's historical transaction preferences, risk tolerance, and investment horizon requirements, providing a comprehensive matching degree score and reasons for the recommendation.

[0039] AI decision-making includes: Multi-round dialogue decision-making process: Account managers interact with the system via a web interface. The system first presents initial marketing strategy suggestions, and account managers can request adjustments, such as "increasing yield requirements" or "reducing risk levels." The system understands the needs through natural language processing, adjusts strategy parameters in real time, and provides a new strategy solution within 5 seconds. The entire process supports up to 10 rounds of dialogue adjustments.

[0040] Strategy Performance Evaluation: The system provides a strategy simulation function, where account managers can input different market scenarios. Based on historical data and machine learning models, the system predicts the expected performance of the strategy under different market environments, including expected return, customer response rate, risk indicators, etc.

[0041] The expert knowledge base has been updated, including: Automatic case archiving: After a marketing campaign ends, the system automatically collects campaign performance data, including actual return on investment, customer response, complaint rate, and other metrics. If the campaign performance meets preset excellent standards (such as return on investment exceeding expectations and customer satisfaction exceeding a threshold), the system will automatically archive the case to the expert knowledge base.

[0042] Intelligent knowledge base recommendation: When account managers develop new marketing strategies, the system automatically retrieves similar success stories from the expert knowledge base and recommends relevant strategy parameters and precautions. The system supports searches based on multiple features, including customer type, product category, market environment, and risk appetite.

[0043] An AI data analysis method based on banking customer marketing includes: Customer information is acquired from multiple data sources, including internal business systems, publicly available external data, and data from third-party service providers, through a data acquisition module. The acquired data undergoes cleaning and preprocessing, including missing value imputation, outlier handling, and data standardization. A customer value score is calculated using an intelligent analysis module, employing the following formula: D i W represents the score of the i-th information item. i This indicates the corresponding weight.

[0044] The intelligent analysis module calculates the customer-product matching score, taking into account customer characteristics, product characteristics, and historical behavior data. The AI ​​decision-making module generates an initial marketing strategy based on the calculation results and supports strategy adjustment and optimization through a multi-round dialogue mechanism. Successful marketing case studies are stored in an expert knowledge base, providing reference suggestions for new marketing decisions.

[0045] The data acquisition module specifically includes: an internal data acquisition unit, used to periodically extract basic customer information and transaction data from the enterprise's core business system; The external data acquisition unit obtains business information and public opinion data through API interfaces and web crawlers; the third-party data access unit receives and processes credit rating and risk data provided by third-party service providers; and the data quality inspection unit verifies the completeness, accuracy, and consistency of the collected data.

[0046] The intelligent analysis module also includes: Customer segmentation unit: Based on customer value score, customers are divided into key customers, potential customers, and ordinary customers; The product recommendation unit generates a personalized product recommendation list based on the customer's matching score with the product. The risk assessment unit calculates the marketing risk level by combining customer credit ratings, transaction behavior, and public opinion information.

[0047] The AI ​​decision-making module specifically includes: The strategy generation unit generates an initial marketing strategy based on the analysis results, including a target customer list, product mix, and budget allocation; The dialogue management unit supports receiving user adjustment instructions via natural language interaction; The strategy optimization unit uses machine learning algorithms to optimize strategy parameters in real time. The effect prediction unit predicts the effectiveness of strategy implementation based on historical data and market conditions.

[0048] The expert knowledge base specifically includes: The case study storage unit is used to store successful marketing cases and related parameters; The feature extraction unit extracts key features from the case and generates feature labels; The retrieval and recommendation unit supports case retrieval and recommendation based on multi-dimensional features; The knowledge update unit continuously updates the knowledge base content based on new marketing results.

[0049] The performance evaluation module is used to collect data on the actual performance of marketing campaigns; compare the differences between expected goals and actual results; analyze the reasons for the differences and propose improvement suggestions; and feed the evaluation results back to the expert knowledge base.

[0050] The multi-turn dialogue mechanism specifically includes: supporting the adjustment of strategy parameters in natural language; real-time display of expected changes in the effect after strategy adjustment; providing simulation of strategy effect under various market environments; recording dialogue history and supporting strategy version backtracking.

[0051] The system implementing the method includes: a data acquisition server for performing data acquisition and preprocessing tasks; an analysis and computing server for running intelligent analysis algorithms; a decision reasoning server for supporting the AI ​​decision-making process; a knowledge base server for storing and managing expert knowledge; and a web application server for providing a user interface.

[0052] Beneficial effects: Enables personalized marketing tool recommendations and configurations, leveraging intelligent analysis to dynamically recommend and configure marketing tools for users, thereby improving the targeting and effectiveness of marketing campaigns and reducing marketing costs; Improve the accuracy of customer identification and product matching, and support the implementation of customized marketing strategies for different customers; By integrating multiple stages such as data preprocessing, precision marketing, and strategy recommendation, a complete data analysis and decision-making closed loop is formed, which helps enterprises improve the accuracy of analysis, the efficiency of decision-making, and the speed of market response. Through continuous learning and updating of the expert knowledge base, the system possesses adaptive optimization capabilities, enabling it to adapt to changes in the market environment.

[0053] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An AI data analysis method based on banking customer marketing, characterized in that, The data analysis methods include: Customer information is collected from multiple data sources to obtain raw data, which is then cleaned and integrated. By assigning weights and dynamically adjusting multi-dimensional customer information, and combining algorithm models to calculate the customer's marketing value score and customer-product matching score, we can assist in achieving precise marketing positioning. An initial marketing strategy is generated based on the calculation results, and the strategy can be adjusted and optimized through a multi-round dialogue mechanism to determine the optimal marketing plan. Successful marketing case studies are stored in an expert knowledge base, providing reference suggestions for new marketing decisions.

2. The AI ​​data analysis method based on banking customer marketing according to claim 1, characterized in that, The customer information includes basic customer information, transaction information, business registration information, public opinion information, and risk information.

3. The AI ​​data analysis method based on banking customer marketing according to claim 1, characterized in that, The process of collecting customer information from multiple data sources, obtaining raw data, and cleaning and integrating the raw data specifically includes: Customer information is obtained from multiple data sources, including internal business systems, publicly available external data, and data from third-party service providers; The collected data is cleaned and preprocessed, including missing value imputation, outlier handling, and data standardization.

4. The AI ​​data analysis method based on banking customer marketing according to claim 3, characterized in that, The acquisition of customer information from multiple data sources, including internal business systems, publicly available external data, and data from third-party service providers, specifically includes: Internal customer data acquisition: Obtain customer basic information tables, account information tables, and transaction records from the enterprise's core business systems; extract data daily using ETL tools, imput missing values ​​using the mean imputation method, and identify and correct outliers using the 3σ criterion; after data cleaning, load the data into the customer subject model of the data warehouse. External publicly available data processing: Obtain business registration information data daily via API interface, including company registered capital, business scope, and shareholder information; We collect public opinion information hourly using web crawlers, including customer-related public opinion data from news, social media, and forum channels; all external data is cleaned, transformed, and then stored in the corresponding data model. Third-party service provider information processing: Receives customer credit ratings and credit limits from third-party institutions daily through the data exchange platform.

5. The AI ​​data analysis method based on banking customer marketing according to claim 1, characterized in that, The process of weighting and dynamically adjusting multi-dimensional customer information, combined with an algorithm model, to calculate the customer's marketing value score and customer-product matching score specifically includes: Calculate customer value score: Among them, D i W represents the score of the i-th information item. i Indicates the corresponding weight; Customer value calculation: The system automatically performs customer value calculation tasks every day at midnight. First, it obtains the latest customer information from various data models, then calculates the score for each customer according to the preset scoring rules, and finally generates a customer value analysis report according to industry and region dimensions. Product matching degree calculation: When an account manager initiates a product recommendation request, the matching degree between the specified customer and the available products is calculated in real time. During the calculation process, factors such as the customer's historical transaction preferences, risk tolerance, and investment period requirements are taken into account, and a matching degree score and recommendation reasons are given in combination.

6. The AI ​​data analysis method based on banking customer marketing according to claim 1, characterized in that, The process of generating an initial marketing strategy based on the calculation results, and supporting strategy adjustment and optimization through a multi-round dialogue mechanism to determine the optimal marketing plan, specifically includes: Multi-round dialogue decision-making process: Account managers interact with the system through a web interface; The initial marketing strategy proposals were presented, and the account manager requested adjustments. By understanding the requirements through natural language processing, adjusting the strategy parameters in real time, and providing a new strategy solution within 5 seconds; Strategy Effectiveness Evaluation: Offers strategy simulation functionality, allowing account managers to input different market scenarios and predict the expected performance of strategies under various market conditions based on historical data and machine learning models. According to claim 6, the AI ​​data analysis method based on banking customer marketing is characterized in that the expected results include expected rate of return, customer response rate, and risk indicators.

7. The AI ​​data analysis method based on banking customer marketing according to claim 1, characterized in that, The provision of successful marketing case studies stored in an expert knowledge base, and the offering of reference suggestions for new marketing decisions, specifically includes: Automatic case archiving: The system automatically collects campaign performance data after the marketing campaign ends; If the activity achieves the preset excellent standard, the case will be automatically archived in the expert knowledge base; Knowledge Base Intelligent Recommendation: When account managers develop new marketing strategies, the system automatically retrieves similar success stories from the expert knowledge base and recommends relevant strategy parameters and precautions. It supports searches based on multiple features, including customer type, product category, market environment, and risk preference.

8. The AI ​​data analysis method based on banking customer marketing according to claim 8, characterized in that, The activity performance data includes actual return on investment, customer response, and complaint rate.

9. An AI data analysis system based on banking customer marketing, employing the AI ​​data analysis method based on banking customer marketing as described in any one of claims 1-9, characterized in that, The data analysis system includes: The data acquisition module is used to collect customer information from multiple data sources, obtain raw data, and clean and integrate the raw data. The intelligent analysis module is used to assign weights and dynamically adjust customer information across multiple dimensions, and combine algorithm models to calculate the customer's marketing value score and customer-product matching score, thereby assisting in achieving precise marketing positioning. The AI ​​decision-making module is used to generate initial marketing strategies based on calculation results, and supports strategy adjustment and optimization through multi-round dialogue mechanisms to determine the optimal marketing plan; The expert knowledge base module is used to store successful marketing cases and provide reference suggestions for new marketing decisions.

Images