False reputation in online transactions: An empirical study

Abstract

Due to the increasing importance of trust in online transactions, the seller reputation system has become a widely used method for creating trust among the transactors and providing incentives for good behaviors in online transactions. A robust reputation system is considered crucial for building a better marketplace for various types of e-commerce. However, false reputation seriously affect the effectiveness of online reputation systems. The purpose of this paper is to explore the causes of false reputation in online transactions through an empirical study. A set of hypotheses are proposed, concerning possible causes for the false reputation creation. Methodology based on data mining techniques is presented to check the validity of the hypotheses. Sales data of women's clothes from TaoBao is used for our case study. Our results indicate several possible problems in terms of the rating system, which may lead to the creation of false reputation.