FinTRACE AI Tool Transforms Financial and Medical Data Analysis

A new AI tool called FinTRACE is set to change how financial and medical data is analysed. Developed by researchers at Sberbank’s Practical Artificial Intelligence Center and the Sber AI team, it turns transaction histories and event-based records into structured knowledge. The system aims to make predictions more accurate while keeping decisions fair and transparent.

A scientific paper on FinTRACE, led by Maksim Makarenko, has now been accepted for presentation at the 49th ACM SIGIR conference, highlighting its growing recognition in the tech community.

FinTRACE works by converting raw financial data into a detailed knowledge base that tracks individual habits and behaviours. Unlike traditional models, it identifies stable patterns before making decisions. This allows banks to spot potential client difficulties earlier and offer support at the right time.

The tool isn’t limited to finance. It can be applied as a universal layer over any event-based data, from predicting customer churn to improving credit scoring. Major banks and digital platforms can also use it to build more flexible and cost-effective AI systems. Beyond finance, FinTRACE has shown promise in healthcare by analysing doctor visits and lab results to detect diseases sooner and plan treatments more effectively. What sets FinTRACE apart is its combination of information retrieval, explainable rules, and large language model reasoning. By integrating these with structured temporal data, the system ensures decisions are not only precise but also easier to understand. This approach could make financial services more reliable while reducing bias in risk assessments and personalised offers.

FinTRACE is already gaining traction, with its research set to be presented at a leading AI conference. Banks and healthcare providers could soon use the tool to improve decision-making across different sectors. The system’s ability to work with various types of event-based data suggests it may have even broader applications in the future.