AirBnB Price Prediction

By harnessing the power of predictive modeling techniques, we can unlock insights into pricing dynamics that facilitate informed decision-making for hosts and guests alike, ultimately enhancing the Airbnb experience for all stakeholders.

Data source

The dataset used in this project was sourced from Kaggle and comprises 29 features and over 74,000 rows. The dataset encompassed numerical, categorical, text, and boolean features, providing a rich array of information for analysis. https://www.kaggle.com/datasets/stevezhenghp/airbnb-price-prediction/data

Preprocessing

• Identifying and handling missing values

• Handling Outliers

• Label Encoding

Dimensionality Reduction and Feature Selection

• Correlation Analysis

• Wrapper Methods (Recursive Feature Elimination)

• Regularization (Lasso)

Regression Models

• Linear Regression

• Lasso Regression

• Ridge Regression

• Elastic Net Regression

• Support Vector Regression (SVR)

• Random Forest

• Gradient Boosting

• XGBoost (Extreme Gradient Boosting)

• LightGBM (Light Gradient Boosting Machine)

• CatBoost

• K-nearest Neighbors (KNN)

• Decision Tree Regression

• Bayesian Linear Regression

Technologies Used

• Pandas

• Numpy

• Matplotlib

• Seaborn

• Scipy

• Scikit learn

• Jupyter Notebook

Explainable AI

LIME and SHAP