Information Technology Reference
In-Depth Information
Part II
Major Approaches to Learning to Rank
In this part, we will introduce three major approaches to learning to rank, i.e., the
pointwise, pairwise, and listwise approaches. For each approach, we will present the
basic framework, give example algorithms, and discuss its advantages and disadvan-
tages. We will then introduce the investigations of the relationship between the loss
functions used in these approaches and widely used evaluation measures for infor-
mation retrieval, to better understand whether it is guaranteed that these approaches
can effectively optimize evaluation measures for information retrieval.
After reading this part, the readers are expected to have a big picture of the ma-
jor approaches to learning to rank, and to know how to select a suitable approach/
algorithm for their given problem settings.
