Authentication in e-commerce

In online marketplaces like the Internet, both buyers and sellers face high levels of uncertainty. Concerns about security, trust, authentication, fraud, and risk of loss are often cited as among the most significant barriers to the growth of e-commerce [4]. A particularly important factor feeding the uncertainty is that traditional authentication mechanisms based on physical inspection are not feasible online. Simply automating traditional processes used in the physical marketplace cannot solve the authentication problem in e-commerce. Online authentication involves much more than the obvious identification and validation problem. We present here a comprehensive framework for online authentication which has several important practical benefits. For instance, we show that there is a significant temporal aspect to the authentication problem over the lifetime of transactional relationships. Also, we illustrate how the framework can be used to evaluate existing authentication mechanisms. This, in turn, can facilitate the development of new mechanisms and processes to better authenticate the parties, products, and processes involved in online transactions. The issue of authentication has been addressed in a variety of contexts, from secure and distributed computing to mobile systems, as well as e-commerce and autonomous computing. In most of the e-commerce literature, discussion of authentication is limited to identification and identity validation [4, 11]. Recent empirical studies on consumer attitudes towards online stores indicates that trust issues going beyond identification are of significant concern, and furthermore, that these factors are influenced by a variety of demographic and cultural factors in addition to site functionality [6, 7, 8]. Also, several models have been proposed recently for trust mechanisms for autonomous computing environments using software agents [1]. However, systematic research to build a comprehensive body of knowledge in this area is still needed.