A Machine Learning-Based Approach for Crop Price Prediction

Abstract

Agriculture is associated with the production of essential food crops for decades and is the one which is playing an important role in the economy of a country as well as in life of an individual. Due to various uncertain variations in the climatic conditions such as rain and other affecting factors, crop prices vary in an unusual pattern. This variation of prices without the knowledge of the farmer may lead to losses in the economy of the individual who is involved in agriculture. In this paper, we have discussed a well-designed system which accurately predicts the crop prices of future months. We have used a Supervised Machine Learning algorithm that is Decision Tree Regression technique for the design of the prediction model as the data is of continuous form. The parameters, which are considered in the dataset, include crop name, month, year, rainfall and wholesale price index (WPI). We have considered the data of 22 crops in total with 4 parameters. We have developed a user-friendly user interface consisting of 22 crop profiles with the predicted prices. Our results show that the regression model achieved an accuracy of 97.32% which will help the farmer on decision of future crop selection for the growth and also hyperinflation can be avoided.