End-to-end Machine Learning project designed to identify and explain the key drivers of Airbnb listing prices using predictive modeling and advanced interpretability techniques.

📊 Executive Summary

This project develops a pricing intelligence model to understand the key factors driving Airbnb listing prices.

Key Insights:

• Property capacity (bedrooms and bathrooms) is the strongest driver of price
• Host activity and recent reviews significantly influence pricing
• Location has a moderate impact compared to structural features
• Pricing behavior is highly non-linear, making tree-based models more effective

Model Performance:

• Linear Regression (baseline): R² = -2
• Random Forest: R² ≈ 0.5

The analysis highlights the importance of moving beyond prediction toward explainable and actionable insights.

🎯 Business Problem

In competitive short-term rental markets, pricing optimization is critical for:

• Revenue maximization
• Competitive positioning
• Property investment decisions
• Host performance benchmarking

The objective of this project was not only to predict prices, but to understand what drives them.

📂 Dataset