Aggregating ranking methods for information retrieval: an overview
Martin J. J. Vesely, Martin Rajman · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · 2007
In the present report we give details on the work that has been done in the scope of the d-Rank project within its first year run, covering the definition of the aggregated ranking model and a set-up of a test collection of scientific documents. We have based our work on a thorough study of the state-of-the-art in document ranking and related domains, focusing on document rank aggregation in a view of an application in a distributed search engine scenario. We have completed the study of the current state-of-the-art covering the principal approaches of a score-based and order-based rank aggregation and presented the most notable aggregation methods used in distributed search engines.