AdaRank: A Boosting Algorithm for Information Retrieval [ Source: Association for Computing Machinery ]
January 2008- This white paper of Association for Computing Machinery explains a novel algorithm for learning ranking models in document retrieval, referred to as AdaRank. In contrast to existing methods, AdaRank optimizes a loss function that is directly defined on the performance measures. It employs a boosting technique in ranking model learning. AdaRank offers several advantages: ease of implementation, theoretical soundness, efficiency in training, and high accuracy in ranking.