Real time recursive preference learning to rank from data stream
Leonid M. Lyubchyk, Galyna Grinberg · 2016
Dynamic ranking learning problem is considered when the training sample is a data stream, consisting of a sequence of a series of objects characterized by a set of features and relative ranks within each series. The problem is reduced to preference learning to rank on clusters in the feature space of ranked objects, while aggregated training dataset is formed from the centers of clusters and estimates of the average rank of the objects from the cluster. Real-time preference function identification algorithm is proposed based on training data stream includes successive estimates of cluster parameter as well as average cluster ranks updating and recurrent kernel-based nonparametric estimation of preference model.