Medical data analysis and transaction type prediction using machine learning and blockchain technology

Abstract

In the world of healthcare, joining machine learning with block chain tech offers a smart path for future predictions. The study aims on guessing what kinds of transactions happen in healthcare data stored on a block chain. Machine learning is used to sort these transactions right. This helps make healthcare tasks work on their own and do better. Health data is taken from block chains, looked at it closely, and ran various algorithms on it. Using features such as operation, date, and symbolic indicators, logistic regression, decision tree, random forest, and support vector machine are applied to classify transaction types. The decision tree algorithm achieved the highest accuracy at 89.29%, followed by random forest at 67.86%, logistic regression at 33.93%, and support vector machine at 39.29%. The findings demonstrate the effectiveness of machine learning in improving transaction classification within secure, decentralized medical data environments.