Demand Forecasting for Restocking Goods Using Machine Learning

Abstract

In the financial operations, many factors play a decisive role. Price and demand play an essential role among them, since they are the key determinants of the financial activities. Demand is not statically placed. In high-range prices, it is marked by unpredictable fluctuations. The principal determinant of market volatility is this form of fluctuation. We now have intelligent machines that can find the lessons from data in this age of artificial intelligence. Data insights can be obtained using machine learning techniques for prediction purposes. Prediction can be a successful way of eliminating market uncertainty. We try to find techniques for the machine learning in our work to help us predict the future demand for products at any business. Our work is based on the raw data from the website of Kaggle. Machine Learning has various prediction algorithms. We use gradient boosting, neural networking (MLP regression), linear regression, SVM, Decision Tree, regression random, forest regression to find the solution. In order to achieve the optimum accurately, we have compared the accuracy in terms of efficiency