Recently Fazli Can and Bilkent Information Retrieval Group created (Milliyet Collection) the first large-scale Turkish IR test collection and conducted first experiments on it.
At SIGIR ’06, they examined the effectiveness of four different stemming techniques with
eight query-document matching functions [4]. At SIGIR ’07 they presented cluster-based retrieval (CBR) experiments on the same test collection [2]. They made an important, positive impact on comparable Turkish IR research by making this TREC-like test collection available to the other researchers [5].