Model-Based AI Agent for Iris Flower Classification

This project demonstrates a Model-Based AI Agent designed to classify Iris flowers into three categories: Setosa, Versicolor, and Virginica.

The agent relies on rule-based decision logic using petal length as the main feature, simulating how an intelligent system can make decisions based on an internal model of the environment.

To evaluate performance, the agent was compared with Machine Learning models including Decision Tree and XGBoost. The project also includes data visualization such as confusion matrix, feature distributions, and decision boundaries to better understand model behavior.

This work highlights the difference between rule-based systems and data-driven models, showing how simple logic can perform well on structured datasets while machine learning models provide more flexibility for complex patterns.