Advanced Data and Prediction Model for Optimal Soil Fertility Management for Precision Agriculture

Through the exploitation of scientific knowledge and technology, man has made wonderous strides in the field of automation, the focus of which lies in ‘Robotics and Machine Learning’. All around us we see machines taking over work with accuracy and ease. Seen as a subset of artificial intelligence, machine learning relies on data, patterns in data and inference to aid technology in thinking for itself. This paper aims to apply the science of machine learning in the field of agriculture, by carrying soil fertility analysis using most accurate algorithm. The fertility of soil plays a principal role in determining the suitability of cultivating a particular crop on a given soil type. Analysis is carried out by the examination of various properties of the soil like the pH value, Electrical Conductivity, Moisture content, Temperature and (N)Nitrogen (P)Phosphorous (K) Potassium levels, followed up by soil type classification. Finally, a recommendation for the most suitable crop is provided in real time.