Extracting Unstructured Information from the WWW to Support Merchant Existence in eCommerce

In [KM02]a model has been developed to support trust in eCommerce. The model is composed of four main modules where each module is a set of factors the consumer is looking for to trust a virtual merchant. These four modules represent the merchant existence, affiliation, policy and performance. In this paper, we present a model to implement the existence module by developing an information extraction system which aims at localising the required information on the merchant’s website. The system is based on rules that reflect the different ways the information is represented on the websites, their structures and the layout of the websites. The extracted information is then stored in a database to be used for a future evaluation of the trust associated with the merchant website.