Predicting agri-food quality across space: A Machine Learning model for the acknowledgment of Geographical Indications

Geographical Indications (GIs), as Protected Designation of Origin (PDO)and Indication (PGI), offer a unique protection scheme to preserve high-quality agri-food productions and support sustainable rural development at the territorial level. However, not all areas with traditional products are acknowledged GI. Examining Italian wine sector by geo-referenced database machine learning framework, we show that municipalities which obtain GI within subsequent 10 year period (2002–2011) can be predicted using large set (lagged) municipality-level data (1981–2001). We find Random Forest algorithm is best model make out-of-sample predictions GIs. Results there sort optimal condition characterized successful matching wine-growing profession (vineyards), local actors involved (number farmers), physical dimension farms (middle farms). Being in vital economic system distance from major urban centers also emerges among main relevant features predicting success The methodology adopted evidence provided lead policy reflections, light future Common Agricultural Policy (CAP) programming scheduled reform GI’s quality scheme.