A Novel Data Science Model using Supervised Learning for Diagnosing Pesticide Poisoning in Rural Workers

In this project, ML algorithms are employed for diagnosing pesticide poisoning in rural workers. The Decision Tree algorithm is a popular choice due to its simplicity and interpretability. It functions by dividing the data into subsets according to feature values, forming a tree-like structure that can predict outcomes. This model is highly effective in identifying key variables that contribute to pesticide poisoning, as it handles both numerical and categorical data well. On the other hand, the RF algorithm, an ensemble method, builds multiple DT and combines their predictions to improve accuracy and reduce overfitting. By aggregating the results from various trees, RF offers enhanced prediction performance and stability. It is particularly useful in handling complex datasets with many features, offering a more robust and reliable model for diagnosing pesticide poisoning. Both algorithms contribute to the development of an effective diagnostic tool for the health risks faced by rural workers exposed to pesticides.