Solving Market Uncertainty By Predicting Potato Price In Bangladesh Using Regression Techniques

Abstract

Market uncertainty is a continuing problem in Bangladesh. As a result, the prices of our common materials fluctuate a lot. It has a significant impact on the components we use every day. In Bangladesh, potato is the third most commonly cultivated crop. In Bangladesh, it is served as the main meal. Bangladesh is a developing country. Potatoes are the third most popular vegetable in Bangladesh, after rice and wheat, with low-income individuals eating more potatoes than other vegetables. The price of potato affects whether people would eat or go hungry. In this era of artificial intelligence, we now have advance software that can extract information from data. Machine Learning is currently quite popular for predicting this sort of unpredictable fluctuation. We created our dataset using information obtained from Bangladesh's Ministry of Agriculture. We used six typical regression techniques to estimate the price of potato. We used Random Forest Regressor (RFR), Decision Tree, Gradient Boosting, Lasso regressor, Linear Regression and Neural Network Regressor models to predict the daily potato price. All of the models we've created yield results that are very satisfactory. Among all the models we created, the Random Forest Regressor (RFR) produced the best results in all stages