Published on:
8 May 2024
Primary Category:
Applications
Paper Authors:
Vladimír Holý,
Ondřej Sokol,
Michal Černý
Proposes purchase-based method to cluster retail products
Formulates as optimization problem and solves via genetic algorithm
Minimizes products from same cluster co-occurring in baskets
Verification tests on synthetic & real drugstore market basket data
Reveals product subcategories beyond expert definitions
Clustering Retail Products by Customer Behavior
This paper proposes a new method to categorize retail products based solely on customer purchase data. It formulates product clustering as an optimization problem, using a genetic algorithm on market basket data to assign products to a given number of clusters. Key assumptions are that customers generally purchase one product per category, and products frequently purchased together should not be clustered together. Tests on simulated and real drugstore data from Czechia demonstrate the method can identify categories similar to those defined by experts, as well as revealing potential subcategories. The technique could help optimize shelf layouts, promotions, and product substitutions when stock runs out.
Unified algorithms for dynamic assortment planning under MNL choice model
Neural networks predict customer purchases from offered product sets
Simulating customer shopping behavior to evaluate retail AI systems
Location decisions for product warehouses to maximize customer satisfaction
Customer segmentation in UK retail using clustering algorithms
Forecasting e-commerce sales
No comments yet, be the first to start the conversation...
Sign up to comment on this paper