A new and promising approach to document clustering consists of utilizing previously formed clusters of queries to cluster documents. To employ this approach in practice a similarity measure for queries must be available. This requirement does not cause any problem in the case of information...