Garment Sales Prediction: A Machine Learning Approach

Abstract

This paper presents a method based on machine learning to predict garment sales within a retail environment. To this end, the paper proposes an optimization system of the stock levels of retailers through better-informed decisions by using improvements in revenue realization employing such methodology. The objective of this research is to establish an accurate predictive model for the sales of garments taking into account past sales, seasonality and other influences like economic conditions and garment Industry. We test several machine learning techniques such as regression analysis, time series, and ensemble techniques to predict how far we can go to categorize and predict sales on individual products, multiple products and product categories. Several algorithms that had been tested and compared, were considered for their efficacy and accuracy on predicting sales trends.