Airbnb users 1st booking destination prediction - Kaggle

Context

New users on Airbnb can book a place to stay in 34,000+ cities across 190+ countries. By accurately predicting where a new user will book their first travel experience, Airbnb can share more personalized content with their community, decrease the average time to first booking, and better forecast demand.

Objectives

Primary: Predict in which country new US Airbnb users will make their 1st booking based on their demographics, web session records and some summary statistics.

Content

• Important Data preparation required to feed model with good features
  • EDA
  • Data cleaning
  • Feature engineering
  • One-hot-encoding for categorical variables

• ML models
  • DecisionTreeClassifier
  • LogisticRegression
  • RandomForestClassifier

**Overall relatively poor performance even with accuracy close to 70%. Models predicting Mostly NDF and US and almost never other destinations. Better feature engineering as well as hyperparameters tuning would help.

For more details see GitHub ML projects - Airbnb users 1st booking destination prediction - Gregoire LAURENT.