[en] Incorrect tagging of proper nouns in disambigation.xml (NNP_DER_SPIEGEL and others)

[MikeUnwalla]

I think that DER spiegel is a German newspaper.

Java rule EN_SPECIFIC_CASE finds the typo, but 'DER spiegel' is not a proper noun, and should not get the postag NNP + NNP.

Who or what is the Seven NATion Army?

NNP_SEVEN_NATION_ARMY gives NATion a postag.

THE CITY OF THOUSAND OAKS IS IN CALIFORNIA.

NNP_THOUSAND_OAKS does not give NNP + NNP to 'THOUSAND OAKS'.

Suggestion: move most multiword proper nouns from disambigation.xml to multiwords.txt. The logic of multiwords.txt makes sure that incorrect typography will not get a postag and that upper-case text will get a postag.