Simple XQuery expressions to extract useful information from some real XML data.
The datafiles are from the British National Corpus, a collection of English language data widely used by linguists and computational linguists, together with a schema. There is a full explanation of the schema1, but the important parts here are:
--> "s" represents a sentence, containing possibly many children representing words in linear order;
--> "w" YYY "/w" represents word YYY.