Learning to Rank with Ensemble Ranking SVM

摘要

In this paper, we propose a novel learning to rank method using Ensemble Ranking SVM. Ensemble Ranking SVM is based on Ranking SVM which has been commonly used for learning to rank. The basic idea of Ranking SVM is to formulate the problem of learning to rank as that of binary classification on instance pairs. In Ranking SVM, the training time of generating a train model grows exponentially as the training data set increases in size. To solve this problem and improve the ranking accuracy, we introduce ensemble learning into Ranking SVM. Therefore, Ensemble Ranking SVM remarkably improves the efficiency of the model training as well as achieves high ranking accuracy. Experimental results demonstrate that the performance of Ensemble Ranking SVM is quite impressive from the viewpoints of ranking accuracy and training time.