Identifying Popular Products at an Early Stage of Sales Season for the Apparel Industry

This paper proposes a new indicator, AW Sales (Average Weekly Sales in Main Sales Period), to measure product popularity unaffected by differences in store traffic, number of stores with initial stock, discounts, and product launch duration.

We developed novel features from weekly sales volume sequences and built a product popularity classification model using ranking algorithms, pioneering their application in sales prediction. The model achieved 78.9% prediction accuracy, identifying fast-selling products 17 days earlier than traditional methods.

Download paper here