Airbnb New User Bookings

Where will a new guest book their first travel experience? You can find the link here.

Team Member

• Brooke Kennedy
• Jamie Chen
• Johnny Chiu
• Veronica Hsieh

Business problem

The goal is to help Airbnb predict which country a new user will make his or her first booking. 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.

This type of classification problem is applicable for many internet/service companies that are looking to drive new users to their website and improve personalization of content to users.

Data Structure

Data for this project mainly includes a users and sessions dataset.

• Users - demographic information about the user along with information about when and how the individual first signed up to be an Airbnb member.
• Sessions - web sessions logs for users. Fields include user actions, session time, and device type for the session

Documentation

• eda.ipynb Jupyter Notebook that contains exploratory data analysis [nbviewer]
• Airbnb 2.26.18 - Data Cleansing & Feature Creation.ipynb Data Cleansing & Feature Creation [nbviewer]
• Logistic.ipynb Jupyter Notebook that contains logistic regression model building. [nbviewer]
• Model-SVM.ipynb Jupyter Notebook that contains SVM model building. [nbviewer]
• Model-GBM.ipynb Jupyter Notebook that contains boosted tree model building. [nbviewer]
• Model-RF.ipynb Jupyter Notebook that contains random forest model building. [nbviewer]
• Model-NN.ipynb Jupyter Notebook that contains neural network model building. [nbviewer]

Directory Structure

project
|   README.md
|   .gitignore
|
|__ data: contains the data downloaded from kaggle
|
|__ script: Source code used for this project
|
|__ scr: notebook related files
|
|__ submission: a folder contains the submission file generated from python script