Metadata Interoperability and Meta-search on the Web

Abstract

Several initiatives for establishing standards for metadata models are being carried out at the moment, but everyone focuses on their own requirements when defining metadata attributes, their possible values and the relation between them. From the point of view of someone who wants to seek and buy information (multimedia content in general) in different environments, this is a real problem, because he has to face different metadata sets, and so, must have different tools in order to deal with them. In this paper, we present a model for the interoperability of different metadata communities, where neither the providers nor the users have to be aware that they all may be working with different metadata models. We are mapping the semantics of different metadata models with the objective of not loosing information when the user and the content provider use different metadata schemas. A "metadata agent" is used to carry out the interoperability functionality. On the other hand, the use of the Internet as a tool for searching information and multimedia content is continuously growing, but the use of metadata in the Word Wide Web is very poor. As a result, many search engines have appeared, that help users to find information. These search engines are able to find information, but generally this information does not follow any metadata standard. Our objective here is to create a meta-search agent able to extract information from the Internet starting from serverindependent queries, which are mapped to search engine specific queries. The results are then reprocessed to provide users with the requested information, again in a server-independent way.