CREAM: A mediator based environment for modeling and accessing distributed information on the web The proliferation of the Internet has led to increasing semantic heterogeneity on the World Wide Web. Networking technology provides physical connectivity, but it does not ensure meaningful data exchange. Therefore, semantic interoperability among heterogeneous information sources continues to pose enormous challenges to the database and other communities. Consequently, cooperation of metadata, ontologies, and mediators is becoming even more important in achieving semantic interoperability; mediators collect information from heterogeneous and autonomous sources and resolve semantic conflicts by referring to metadata and ontologies. Based on the mediator approach, we have implemented a Conflict Resolution Environment for Autonomous Mediation (CREAM) system. The system provides various user groups (e.g., end-users and information integrators) with an integrated and collaborative facility for achieving semantic interoperability among the participating heterogeneous information sources. The detailed description of a theoretical framework for semantic conflict detection and resolution methodology can be found in a paper of A. Gupta et al. (2000). Our major contribution is that our system automatically understands and resolves a large number of semantic and syntactic conflicts among heterogeneous databases. We have tried to minimize the burden on the end users and database administrators while trying to automate the heterogeneous conflict resolution process.