India is a land of agriculture; we are the leading producer of many crops all over the world. It is a major producer of milk, pulses, and jute and is the second largest producer of cotton, sugarcane, vegetables, etc. In today’s world where almost everything is digital, only the agricultural sector has no technological advances. India is still showing a slower pace to adapt to more advanced technologies when it comes to agriculture. Machine learning has emerged as a big data technology and is expected to grow rapidly shortly. With the help of machine learning, we can improve the current situation of the poor pricing model of farmers’ crops and help them by getting the best Minimum Support Price (MSP) for their crops. One of the majorly produce commodities in India which is raw cotton, it has the most volatile price trend as compared to any other crop in India. Due to this volatile nature of the price trend, our farmers suffer the most and the government can also not decide on an appropriate MSP to support them. In this paper, we proposed an approach for time-series data of the market prices of raw cotton to predict the market price trend using different machine learning models. We have used the Facebook Prophet algorithm for a complete year trend that will help our industrialists as well to determine their future ideology and Random Forest regression for day-wise trends. We have considered absolute percentage error as the evaluation metric. By using both Facebook Prophet and Random Forest, we have got the mean absolute percentage error of 0.08 in Facebook Prophet and 0.04 in Random Forest that indicates the efficacy of the proposed models in price trend prediction for cotton crop.