Integration of Heterogeneous Data Models: A Mashup for Electronic Commerce

Abstract

Nowadays, the interoperability between e-commerce sites requires that information must be processed syntactically and semantically. For example, find the best product in two different sites according to different criteria, including numerical and non-numerical attributes (e.g. product name, price, product evaluation, customer opinion, etc.). We present a mashup between the sites eBay and Amazon to search for products according to these criteria. The integration is driven by ontology with a syntactic-semantic process. The result is the best product of the two sites according to criteria (price, user rating, delivery times and geographical location). Additionally, a map is displayed, showing the product information, not only the data related to the product, but also other geographical locations where is possible to buy products and opinions from other clients. Tests showed that including semantic processes in the product search gives better results than when using only syntactic processes.