Hana Bank Evaluates Corporate 'Future Growth Potential' Using Artificial Intelligence

Development of 'Technology-Based ML Model'
Enables Evaluation Based on Owned Technology Beyond Financial Statements

Hana Bank announced on the 8th that, in collaboration with Hana Financial Convergence Technology Institute's ‘Data Modelling Cell,’ it has developed the banking industry's first ‘technology-based ML (Machine Learning) model’ using artificial intelligence (AI) technology to conduct corporate evaluations.

The technology-based ML model primarily utilizes technology-related information such as patents and technology certifications held by companies, technical personnel, current status of technology development, and technology commercialization capabilities as AI training data. This approach moves away from traditional credit evaluations based on financial statements and assesses future growth potential based on the company's current technological assets.

This model quickly analyzes technology evaluation data to automatically identify explainable factors from a risk perspective. To ensure the stability of evaluation results, it is designed to learn from items with low volatility, thereby enhancing the discriminative power of technology indicators. Through this, it can provide customized factors that positively or negatively affect a company's creditworthiness, securing transparency and reliability in evaluations.

An official from Hana Bank's Small and Medium Venture Finance Department stated, “With the development of the technology-based ML model, we expect to offer more financial benefits to small and medium enterprises possessing excellent technological capabilities.” The official added, “Hana Bank will continue to contribute to the growth and development of SMEs through the development of various financial products.”

Meanwhile, Hana Bank is preparing for approval from the Financial Supervisory Service by developing an integrated credit model (Phase 1) that applies highly stable indicators from the technology-based ML model’s evaluation metrics to non-financial credit evaluation items. Once approval is completed, the bank plans to operate the full integrated credit model starting in the first half of 2025.

This content was produced with the assistance of AI translation services.