MARKET-DRIVEN CROP RECOMMENDATION SYSTEM USING PUBLIC DATASETS

Crop selection is critical to determining
the profitability of a given farm. Many farmers use
either traditional knowledge or general advice when
selecting their crops, which do not guarantee they
will receive the highest profits. This project
proposes a machine learning algorithm to sort and
analyze the profits from several different crops by
applying several types of information (soil content
and nutrient levels, climatic conditions, and market
conditions) to create a Random Forest model to
predict the profitability of a set of crops.
The proposed system utilizes the output from its
model to provide suggestions as to which crop will
yield the highest profit. The analysis performed on
the data indicates that by combining data from
agriculture with data from the market, agricultural
producers can enhance their decision-making
capabilities.
This system allows producers to decrease their
exposure to risk, prevent/protect against potential
loss, and increase their income through their crop
selection process. It is a well- supported and
practical solution to modern-day problems in
agriculture.