• data science
• machine learning
• ensemble classifiers
• real estate market
• time to sell
• proptech

The usage of machine learning techniques in the real estate market has become a de facto standard. One of the relevant skills that professionals rely on is artificial intelligence and machine learning in their ability to "predict" the future of the real estate.
This study aims to predict the time of sale of residential properties in Italy using machine learning techniques, and identify the factors that most influence the timing of property sales. After conducting a thorough analysis of the data with regard to temporal and spatial dimensions, three distinct datasets underwent training and testing employing multiple ensemble models for classification, namely Random Forest, LightGBM, XGBoost, and CatBoost. To do so, various approaches were exploited and compared, such as binary classification, multiclass classification, and multilabel classification.