Description

Data Preprocessing

• Loaded dataset using pandas and preprocessed data with sklearn.
• Performed one-hot encoding for categorical variables.

Model Training and Evaluation

• Trained RandomForestClassifier for predicting discontinued medicines.
• Evaluated model accuracy reaching 97%.
• Explored decision tree depth for understanding model complexity.

Visualization

• Visualized decision trees and created scatter plots using Plotly.

Medicine Recommendation Function

• Developed function to recommend alternative medicines based on input.
• Considered composition and availability for accurate suggestions.

Testing and Results

• Tested with "Augmentin 625 Duo Tablet" and generated recommended substitutes.
• Provided details such as price, manufacturer, type, and pack size.

This project integrates machine learning into healthcare applications, providing reliable medicine recommendations based on thorough data analysis.