Microsoft Research at TREC 2011 Web Track
Bodo Billerbeck, Nick Craswell, Dennis Fetterly, Marc Najork · 2011
This paper describes our entry into the TREC 2011 Web track. We extracted and ranked results from the ClueWeb09 corpus using a parallel processing pipeline that avoids the generation of an inverted file. We describe the components of the parallel architecture and the pipeline, how we ran the TREC experiments, and we present effectiveness results. 1