Heterogeneous and Distributed Information Retrieval

In the last few years there have been the explosion in the use of heterogeneous distributed systems. Ranging from simple Network of Workstations to the more modern and complex Grid systems, the adoption of distributed systems instead of massively parallel supercomputers has been preferred due to their reduced cost of ownership. These kinds of systems pose many challenges in terms of information access, storage and retrieval. Usually, in fact, instead of having collections stored at a single site they are collected, and sometimes managed, at different sites (possibly owned by different institutions). Particular interest has been expressed on architectures and specifications for information retrieval in the context of heterogeneous distributed computing systems.

This workshop will focus on new methods and algorithms to efficiently and effectively access data distributed over large heterogeneous distributed systems. The workshop particularly encourages papers that address the creation and the search in distributed, dynamic information systems as well as papers presenting novel architectural solution for these systems. However, more broadly, papers are solicited on any topic related to information retrieval in distributed architectures, including the topics listed below.

List of topics:

• Meta-IR: coping with heterogeneous distributed data
• Algorithms for distributed IR
• Grid IR
• Peer-to-Peer (P2P) IR
• Complex queries specification and resolution
• Collection selection
• Query routing (especially in Grid and P2P systems)
• Data fusion
• Query execution scheduling
• Index structures
• Specific architectures
• Scalability
• Fault tolerance
• Privacy
• Applications

Submissions:

Important Dates