MACHINE LEARNING MODELS FOR THE PREDICTION OF HIGH-RISK SENSITIVE RECORDS: COMPARATIVE STUDY & PERFORMANCE ANALYSIS

Big data has revolutionized the way health risks are identified and evaluated in healthcare. It
efficiently processes and analyzes clinical data with exceptional speed and precision. It facilitates the
collection and analysis of extensive, heterogeneous datasets such as electronic health records (EHRs).ML
models analyze complex healthcare data to identify high-risk patients and the most influential features
for early and prioritized diagnosis. It facilitates timely intervention in potential health emergencies by
collecting and analyzing raw data from sources such as electronic health records and wearable monitoring
devices. The increasing demand for advanced clinical data analytics highlights the importance of
leveraging enhanced big data techniques and validated algorithms to accurately detect high-risk cases
and improve risk prediction across diverse clinical settings. This enhancement integrates diverse Machine
Learning (ML) models including Naive Bayes, K-Nearest Neighbors, Decision trees, Logistic
Regression, Random Forest, Neural Networks and Stochastic Gradient Descent to optimize healthcare
data interpretation. Consequently, the proposed approach outperforms other methods in terms of
classification metrics, including precision, recall, and F1-score with a reduced error rate.