Banking Operation Support Using Customer Action Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing banking operation technologies fail to predict customer actions that contribute to overall management and do not effectively encourage customer actions, while also neglecting the sound management of financial institutions.
Innovation Solution
A banking operation support system utilizing machine learning to predict customer actions and financial transaction conditions, incorporating customer and external states, to optimize balance sheets by controlling deposit and loan conditions, and performing investment and funding simulations to meet regulatory requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine learning is used to predict customer actions and optimize banking operations, then productivity and profitability are improved, but device complexity and data processing requirements increase
Solution Approach 1:
The banking operation support system is divided into distinct functional modules: a prediction unit that uses machine learning to forecast customer actions, a simulation unit that performs investment and funding simulations, and a support generation unit that creates operational recommendations. This segmentation allows each module to specialize in specific tasks, improving overall productivity while managing system complexity through modular design.
Solution Approach 2:
The system introduces an intermediary banking operation support system that acts as a bridge between raw customer data and banking decision-making processes. This intermediary processes and analyzes data through machine learning models, transforming complex data into actionable insights without requiring direct complex interactions between data sources and banking operations.
2Measurement precision
If comprehensive customer data and external states are collected and analyzed, then prediction accuracy is improved, but loss of time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by pre-collecting and organizing customer data and external state information before predictions are needed. Customer information, account information, transaction information, and external state data are maintained in ready-to-use formats, allowing the machine learning models to quickly process information when predictions are required, thus improving prediction accuracy without excessive processing delays.
Solution Approach 2:
The system dynamically adjusts parameters such as the depth of data analysis, the complexity of machine learning models applied, and the scope of simulations performed based on the specific banking operation context. This allows the system to optimize between prediction accuracy and processing time by selecting appropriate parameter levels for different scenarios.
Data Source
AI summary
Provided are a system, a method, and a program of banking operation support for supporting execution of accurate and efficient banking operations. A control unit (21) of a support server (20) acquires external state information, a financial transaction condition, customer state information, and a financial transaction state of the same time period, from a back-end system (40); performs machine learning using a data set made up of the acquired pieces of information as learning data, and generates a customer action prediction model which associates the financial transaction condition with the financial transaction state; and calculates, for a current external state and a current customer state, a financial transaction condition corresponding to a target value of a transaction state using the customer action prediction model.