Comparison of Different Machine Learning Algorithms for Multiple Regression on Black Friday Sales Data

摘要

During the Black Friday sale, all the retail shops are crowded. Most products are marked down with discounts and customers rush in to buy the products. It is difficult for customers to buy the products even with a solid plan. But, the shop owners face even more difficulty on controlling the crowd with limited staff and in targeting prospective customers. Several techniques have been employed to tackle this problem, but they are not that successful. A prediction model is a technique that has proved promising in solving the problem. This study focuses on the field of prediction models to develop an accurate and efficient algorithm to analyze the customer spending in the past and output the future spending of the customers with same features. In this study, different machine learning techniques such as regression and neural network to develop a prediction model are implemented and a comparison is done based on their performance and accuracy of prediction. These techniques are implemented using different algorithms and on different platforms to find the best predication. We implemented seven different machine learning algorithms. Further, this study discusses the data pre-processing and visualization techniques employed to attain the optimal results.