PREDICTING THE PRICE OF COTTON USING RNN AND LSTM

• Main Author: MOHAMAD, AHMAD LUKMAN
• Format: Final Year Project
• Language: English
• Published: IRC 2020
• Subjects: Q Science (General)
• Online Access: http://utpedia.utp.edu.my/21716/1/24617_Ahmad%20Lukman%20Bin%20Mohamad.pdf; http://utpedia.utp.edu.my/21716/

The covid-19 has resulted in the volatile fluctuation of the commodities prices such as the price of cotton and causes a disruption in the supply chain of cotton. This research is aimed to utilize the machine learning algorithm to try and predict the future price of cotton by using the historical price of cotton provided by investing.com. Prediction of the cotton price will enable companies to hedge their position in contract for difference market to minimize the effect of volatile price fluctuation and help to stabilize the supply chain of cotton. The research will be conducted through four stages of machine learning development methodology consisting of data gathering, data processing, model fitting, and performance analysis. Data will be gathered by importing the historical prices of cotton from investing.com into a csv file. The data will then be processed to eliminate the unused columns and to fill in any void cells. The data will then be separated into training set and testing set and will be feed to the machine learning algorithm to find the pattern and try to do prediction. The accuracy of each machine learning algorithm will be recorded and analyze to come up with the best model for the dataset. As of now, the data gathering, and data processing stages has been done and the research shall continue with the model fitting and performance analysis