In this chapter we will introduce the pairwise approach to learning to rank. Specifically we first introduce several example algorithms, whose major differences are in the loss functions. Then we discuss the limitations of these algorithms and present some improvements that enable better ranking performance.
CITATION STYLE
Liu, T.-Y. (2011). The Pairwise Approach. In Learning to Rank for Information Retrieval (pp. 49–70). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-14267-3_3
Mendeley helps you to discover research relevant for your work.