| Abstract |
Nowadays, public and private companies, are in a constant race to increase profitability, chasing the costs reduction while facing the market competition. Also in the agriculture an analysis of cost-effectiveness, measuring technological innovation and profitability becomes necessary. The 'smart farm' model exploits information coming from technologies like sensors, intelligent systems and the Internet of Things (loT) paradigm to understand the intluential and non -influential factors while considering environmental, productive and structural data coming from a large number of sources. The goal of this work is to design and deploy practical tasks that exploit heterogeneous real datasets with the aim to forecast and reconstruct values using and comparing innovative machine learning techniques with more standard ones. The application of these methodologies, in fields that are only apparently refractory to the technology such as the agricultural one, shows that there are ample margins for innovation and investment while supporting requests and needs coming from companies that wish to employ a sustainable and optimized agricultural industrial business. |
