Information Technology Reference
In-Depth Information
Part V
Practical Issues in Learning to Rank
In this part, we will introduce how to apply the learning-to-rank technologies men-
tioned in the previous parts to solve real ranking problems. This task turns out not
to be as trivial as expected. One needs to deal with training data generation, fea-
ture extraction, query and document selection, feature selection, and several other
issues. We will review previous works on these aspects, and give several examples
of applying learning-to-rank technologies.
After reading this part, the readers are expected to understand the practical issues
in using learning-to-rank technologies, and to be able to adopt an existing learning-
to-rank algorithm in their own applications.
