This adds feature extraction for the LTR baseline method.
Not all features are extracted. Many of the "table" features cannot be reproduced as no table page information (except for the page title) is provided in the dataset. There are however some features that can still be added, such as the IDF values for the query terms IDF_f, and the qInPgTitle and qInTableTitle features. The extraction of these features will come in a later PR.
This adds feature extraction for the LTR baseline method.
Not all features are extracted. Many of the "table" features cannot be reproduced as no table page information (except for the page title) is provided in the dataset. There are however some features that can still be added, such as the IDF values for the query terms
IDF_f
, and theqInPgTitle
andqInTableTitle
features. The extraction of these features will come in a later PR.