Temporal expressions, such as between 1994 and 2000, are
frequent across many kinds of documents. Text retrieval,
though, treats them as common terms, thus ignoring their
inherent semantics. For queries with a strong temporal component,
such as U.S. president 1999, this is problematic, since
documents can not be reliably matched to the query.
Our key contribution in this work is a novel approach
that integrates temporal expressions in the language modeling
framework, thus turning them into first-class citizens
of the retrieval model. In addition, we present experimental
evidence showing an improvement of retrieval effectiveness