This R notebook is the result of EDA performed on Airbnb data collated on the 10th of July 2019 for listings in London and aimed to find a model which could predict the price of a given Airbnb listing most accurately.
The modelling approaches used are more "traditional" statistical and machine learning approaches including multi-linear regression and decision trees using techniques such as random forests and bagged trees.
This project suffered from data quality issues that could not be resovled in the time available including rare factor levels that prevented a subset of the data from being used in test and training sets. In future this could resolved by taking a larger a dataset and/or more effectively filtering the data set. However, it was interesting to see the level of accuracy that can achieved without resorting to more complex machine learning algorithms.
For further information on the approach used please see the report that has been committed to this repo.