Unveiling the features of successful eBay smartphone sellers

Abstract

The present study adopts a data mining approach based on support vector machines (SVM) for modeling the number of sales of smartphone devices by eBay sellers. The data-based sensitivity analysis was adopted for extracting meaningful knowledge translated into the relevance of each input feature for the model. Such approach allowed unveiling that the number of items the seller also has on auctions, the price and the variety of products the seller offers are the three features that influence most the number of sales, in a total of almost 25%, surpassing the relevance of the features related to customers' feedback.