- Pandas
- Numpy
- Plotly
- Matplotlib
- Sklearn
- XGBoost
- NLTK
Through our analysis of this data, we explore booking trends on airbnb, find factors affecting the house prices, understand the sentiments of people and create a machine learning model to predict the house prices based on the processed features.
- EDA.ipynb: This jupyter notebook comprises data cleaning and preprocessing, feature engineering and modelling for airbnb house price prediction.
- sentiment_analysis.ipynb: Analysis of user reviews for predicting sentiments using unsupervised machine learning algorithm.
The results of our analysis have been summarised in a blog post on Medium and it can be accessed here.
The licensing for this data can be found here. This code can be used free of charge and I would encourage the reader to extend this work for even better insights.