Find unlexicalised compounds in the dictionaries

We have made a python script that goes through all compound nouns in the dictionary and check whether they are found as lexicalised compounds in the analyser. The script is find-unlexicalized-compounds.py , and it can be found in giella-core/scripts. The script should be in your path. To use The script: Stand in dict-xxx-yyy (here: dict-smn-fin) and collect the unlexicalised smn compounds, as follows:

find-unlexicalized-compounds.py -i src/N_smnfin.xml -l smn -o missing.txt

The resulting nouns in missing.txt may be made into candidates for addition to nouns.lexc with another script missing.py as follows:

cat missing.txt | missing.py -l smn > missing.lexc