Big data analytics and artificial intelligence serving agriculture

Abstract

Agriculture is a very important sector in the Moroccan economy. Its contribute more than 15% of Morocco’s GDP and it’s employs about 45 percent of the total workforce with a similar portion of the population living in rural areas. We find in Morocco three types of agriculture: • First, export-oriented farms, irrigated, modern and highly capitalized. • Second, the agriculture in large-scale irrigated areas. • Third, the Rain-fed agriculture with more and less favorable land in the northeast, south and east. Knowing that the digital transformation is lever of the industrial revolution 4.0 and Big Data and machine learning are huge in predicting things like when you might want to use certain pesticides, we proposed in this paper an advanced intelligent systems for sustainable development applied to agriculture. We used methodological data to preventing the Cereals production rate in an area characterized by an unstable climate using artificial neural networks. We started our process by collecting, preprocessing of methodological of more than 60 years, and we stared a comparative study to find the most efficient neural architecture in terms of complicity and in terms of recognition rate. And in the end, we presented the results and prospects of the system.