Airbnb is one of the fastest growing companies and has been increasing it's user base rapidly. Airbnb’ing continues to expand in popularity among both leisure and business travelers. Airbnb gives independence to the host to price their listing. This provides interesting use cases to solve for both the hosts as well as the guests. I took up two use cases 1) Predicting competitive price for a listing based on the characteristics of the listing which helps the hosts when they enter the market. 2) Forecasting prices of listings based on the multivariate time series data to help the guests plan their trip and accommodation. The tasks performed in implementing the project are: 1. Exhaustive Data Exploration, Data preprocessing and Data Wrangling of the Airbnb dataset. 2. Used Ensemble models such as Random Forest, XGBoost and LightGBM for price prediction. Also used Ridge Regression and KNN. Computed Median Squared Error and Mean Squared Error of the test data. 3. Performed hyperparameter tuning on all 5 models using ARC(A Root Cluster for Research into Scalable Computer Systems) and the results showed a marked improvement. 4. Used Arima (AutoRegressive Integrated Moving Average) and BSTS (Bayesian Structural Time Series) to capture the seasonal effects and trends in the data which helped in forecasting the prices of the listings with great accuracy. - View it on GitHub