Since 2008, guests and hosts have used Airbnb to expand on traveling possibilities and present more unique, personalized way of experiencing the world. The objective of this project is to explore the listing features of this dataset and get to grips with the Airbnb market in the city of Cape Town, South Africa.
Assist new airbnb hosts in determining appropriate pricing / ranges for their listings based on the features or predictor variables in the dataset. Current hosts could use the model to determine if their listing prices are overpriced or under.
- Programming fundamentals and the basics of the Python programming language (Python 3.x), Numpy, Pandas, Matplotlib, Seaborn
- Fundamentals of Supervised Learning: Regression
- Scikit-learn
- Wards closer to Tourist hotspots have more listings.
- Listings further away from tourist attractions are generally more affordable.
- Airbnb is growing it's market share considerably as people search for luxury, comfort and affordability.
- Covid-19 has had an extensive impact on the tourism industy in particular.
- Cape Town is an expensive city, so save up.
- Overall, with the features provided, prediction of the listing price is not reliable with the lowest RMSE being of the order of 2500 ZAR.
- Geographical features seem to be the most influential, which is intuitive as the price of a listing would depend on the location, proximity to tourist attractions, type of neighbourhood/ward.
- Model performance could be improved by obtaining more features relating to the listing aside from the geographical features provided such as the size / area, number of rooms, ammenities available such as WiFi, parking and so forth.
- These are conveniences that likely differentiate between listings in the same wards/neighbourhoods which should help with accuracy.