Any sources for the click model based relabeling mentioned in the readme

[shanmukh98]

Was going through the readme and found this

The click model that is mentioned here seems interesting
any references for the method mentioned here would be great.

[kretes]

If you're asking about some literature sources about click models in LTR in general, then look at Seq2Slate paper https://arxiv.org/pdf/1810.02019.pdf (mentioned in https://github.com/allegro/allRank/blob/master/allrank/click_models/cascade_models.py#L37C51-L37C83) which explains a procedure of using a reasonable click to obtain click-through data programatically.

This click model is implemented as well as some dummy click models like random-click, and as well as some raw ideas about some more complex click-models

[mhsyno]

All the click models implemented are located in click_models module with some short explanation in docstrings. As mentioned in the part of readme you’ve referred to, before applying a click model you need to pass previously trained allRank model and pass it to rank_and_click.py script with its config modified by specifying click model (along with its arguments) you want to apply - that is because some click models (e. g. cascade model) need relevance (output of the model) to decide which item to click.