Supervised Learning: Predicting House Prices Across Italy

💡 The goal of the project is to predict house prices in different Italian cities using the following datasets:

• In train.csv you will find a set of rows, each containing data for one apartment for sale. The target variable that you have to predict is the sale price.
• poi.csv contains the coordinates of points of insterest that you can use to further enrich the features available in your training dataset.

Metric: Submissions were evaluated on Mean-Squared-Error (MSE) between the predicted value and the observed sales price

The file is currently structured as follows:

Introduction

Part 0: setting up environment

Part 1: Data Exploration

└── 1.1 Data Visualization └── 1.2 Data Cleaning │ └── 1.2.1 missing data │ └── 1.2.2 outliers 1.3 Feature Engineering 1.4 Mixing the ingredients Part 2 : Modelling

└── 2.1 splitting thedata └── 2.2 modelling └── 2.3 comparing models Part 3: A-B testing / CV Conclusions