Application of Data Analytics in Agriculture Sector for Soil Health Analysis: Literature Review

Soil is the most important part of agriculture. This paper compares different prediction models for factors which effects on soil health. Soil moisture, nitrogen, phosphorus, potassium, organic matter, heavy metals content is important for farmers to determine how much irrigation is required, which type of crops can be grown in such soil, which fertilizers to use for better yield from soil. High content of heavy metal can degrade the quality of soil. Such type of soil is also less useful for crops. This paper compares Prediction of heavy metals present in soil with SVM, RF, ELM. Among these three RF is found out to be most accurate and stable. SVM gives best accuracy for prediction of soil moisture and predicting the soil nutrients like N, P, K.