November 26, 1999
School of Information Technology and Engineering
University of Ottawa
Using Noun Phrase Heads to Extract Keyphrases;
Using Human Heads to Evaluate Keyphrases
In the talk I will describe a simple system for choosing noun phrases from a document as keyphrases. The system chooses a noun phrase based on its length, its frequency and the frequency of its head noun. Noun phrases are extracted from a text using a trivial base noun phrase skimmer and an off-the-shelf online dictionary.
I will also talk about the results of the recent experiments involving human judges (among them several TAMALEans): the noun phrase-based system seems to perform about as well as Peter Turney's Extractor; ratings for individual keyphrases do not necessarily correlate with ratings for sets of keyphrases for a document; agreement among unbiased judges on the keyphrase rating task is poor.