Value-Aware Product Recommendation by Customer Segmentation using a suitable High-Dimensional Similarity Measure
This approach tackles sparse data by directly incorporating product and user revenue contributions into recommendations.
Deep Dive
A value-aware product recommendation framework encodes revenue contributions into the user-item matrix and uses a suitable high-dimensional similarity measure for customer segmentation. Users are grouped based on the revenue similarity of their purchase baskets, supporting three recommendation strategies: revenue share, product popularity, and expected profit. The method is validated through simulation experiments and a real-world application using the UCI Online Retail dataset.
Key Points
- Encodes revenue contribution of each product and user into the user-item matrix for customer similarity computation
- Uses a novel high-dimensional similarity measure to handle sparsity typical in retail data
- Validated on UCI Online Retail dataset with three recommendation strategies: revenue share, popularity, and profit generation
Why It Matters
For e-commerce and retail, this method turns recommendation engines into profit drivers by segmenting customers based on actual revenue behavior.