claims management

Machine Learning Streamlines Claims Management in Financial Market Commission

The Financial Market Commission created an automated system using machine learning to classify claims and inquiries, resulting in significant improvements in response times and efficiency.


The Challenge

The Financial Market Commission (C.M.F), responsible for ensuring the proper functioning, development, and stability of the financial market, is receiving a high and growing number of claims and inquiries from the public, revealing a high inefficiency in response capacity and consequent delay in processing. This entity sought solutions to address this problem by creating an automated system for the structuring and massive processing of claims and inquiries, improving response times significantly, and strengthening its supervisory duties.

The Solutions

To achieve the goal, we developed a system that uses machine learning techniques to classify the claims received by market and entities, products, and matters claimed. This process began with the collection and cleaning of available data, i.e., claims with their corresponding classifications. Then, we used classification algorithms and neural networks to analyze the information, test alternative models and hyperparameter combinations until achieving the expected results.

Our solution also included the development of a web service to enable interoperability of the proposed system, i.e., to be consumed and integrated by the client entity’s systems. Additionally, we developed scalability elements of the solution through automation and scalability practices known as MLOps, providing capabilities for continuous model retraining and identifying issues that could affect the solution in a production environment.


What we achieved

The results of our solution are highly satisfactory. In tests conducted in relevant and simulated environments, we achieved 95% accuracy in classifying claims by market and 89% accuracy in classifying by entities, products, and matters claimed.

In real-world testing, system users report that response capacity and quality of analysis improve significantly, reducing the average time a claim spends waiting to be classified from a week to less than 5 seconds, resulting in greater efficiency in the management process and financial market supervision in general.