Tree in Itsdb Profile

From here, the result of a parse contains:

derivation :string # derivation tree for this reading
surface :string # surface string (e.g. realization)
tree :string # phrase structure tree (CSLI labels)

  1. How is the label for each node determined in tree? Are they derived one by one from the derivation’s feature structure via certain rules, e.g. HEAD of types partn and noun implies N?
  2. Is there a inventory of CSLI labels with their definitions? I do not quite get the meaning of some complex labels, e.g. VP-C/NP-CJ, LADV, etc.

I am not sure about what keyword I should search with in the wiki. Any link to relevant resources are welcome. Thank you!

Re 1: They are determined mainly from the HEAD feature value and whether some lists in the feature structure are empty or non-empty.

Re 2: For the ERG I think you can look at the parse-nodes.tdl file. I’m not sure if there’s a listing that describes what the labels mean, but there might be some overlap in the ErgRules wiki.

1 Like