Utilizing Random Forest to Diagnose Pneumonia in Children

abstract-Around the world, pneumonia takes the lives of about sixteen percent of children under five years old. This serious respiratory illness targets the lungs, affecting small air sacs called alveoli. As people breathe, these sacs fill with oxygen, but when pneumonia strikes, they fill with pus and fluid, making it hard to breathe and reducing oxygen absorption. In Southeast Asia, India bears a heavy burden, accounting for 36% of the cases in the region identified by the World Health Organization (WHO). Machine learning (ML) classifiers are crucial for analyzing large and complex datasets, helping with tailored treatment plans, predicting disease outcomes, and improving the efficiency of healthcare delivery. This research focuses on using ML techniques like Random Forest, Decision tree, K-Nearest Neighbors to diagnose pneumonia in children. It assesses the accuracy of these algorithms, finding Random Forest to be superior in detecting pediatric pneumonia compared to Decision Trees and KNN.