AirBnB Housing Price Prediction Project

R, Random Forest, XGBoost, K-Nearest Neighbors (KNN)

Project Overview

Predicted housing prices on Airbnb Los Angeles houses dataset by training and benchmarking machine learning models including Random Forest, K-Nearest Neighbors (KNN) in R (random forest, xgboost, knn). I performed log-transformation, stratified cross validation, and hyperparameter tuning on models; yielded 63% R-squared and 54% RMSE on best performing model using Bagging

Installation and Setup

• Technologies: R Studio
• R Version: 4.2.2
• Packages Used:
  • Data Manipulation: tidyverse, tidymodels
  • Data Visualization: ggplot2, yardstick, corrplot, rpart.plot
  • Machine Learning: randomForest, xgboost, vip, ranger, kernlab, kknn, baguette

Data

Source Data

• http://insideairbnb.com/get-the-data/
  • Inside Airbnb is a mission driven project that provides data and advocacy about Airbnb's impact on residential communities. We work towards a vision where data and information empower communities to understand, decide and control the role of renting residential homes to tourists.
• http://data.insideairbnb.com/united-states/ca/los-angeles/2022-09-09/visualisations/listings.csv
  • Summary information and metrics for listings in Los Angeles (good for visualisations).

Data Acquisition

Data Preprocessing

Our variables of focus:

• id —
• id —

Results and Evaluation

Desc

Future Work

Desc