Affinity-driven prediction and ranking of products in online product review sites
CIKM '10 Proceedings of the 19th ACM international conference on Information and knowledge management
| Hi-index | 0.00 |
Online product ranking is complicated by conflicting attributes. The authors propose a Rainbow ranking system to solve this problem in e-commerce. This approach extends eBay's one-to-one shopping paradigm to a flexible one with many-to-many transactions. The system demonstrates a 50 to 70 percent transaction increase over skyline and eBay ranking methods. Buyers are allowed personalized selections of commodity products, whereas sellers benefit from both wholesale and retailing online. The Rainbow scheme thus creates a win-win situation for both buyers and sellers in widened e-commerce marketplaces.