Please, what is the best chapter book that is talking about click models in learning to rank in which it states about the need for offline relevance labels in the click models probabilities?
Please let me know if these references/sites are useful to you:
1. Online Learning to Rank in Stochastic Click Models
https://arxiv.org/abs/1703.02527Mar 7, 2017 ... To overcome this challenge, we propose a novel algorithm, which we call MergeRank, for learning to rank in a class of click models that satisfy ...
2. Online Learning to Rank in Click Models
https://pdfs.semanticscholar.org/1355/4f8fd86a75220f132f199ac5126752af8d7f.pdfMar 7, 2017 ... common framework for learning to rank in a general class of click models, which includes both the CM (Craswell et al., 2008) and the PBM ...
3. Explore Click Models for Search Ranking - Yuchen Zhang
https://zhangyuc.github.io/files/wang10cikm.pdfIn this paper, we focus on learning a ranking function by taking the results from a click model into account. Thus, besides the editorial relevance data arising from ...
Thank you very much. The links do not work. However, I also found this chapter by Tie-Yaun and he mentioned some issues in click models and also he created LETOR datasets (MSLR-WEB10K and MSLR-WEB30K) that include actual user clicks as features and user dwell time feature: