Key words (measuring keyness) get presented in a list with all those meeting the criteria ordered by how well they meet them, this may be arguably not a simply ranked list. Key words with equal keyness scores vary in important ways such as their frequency in the corpus, number of texts etc. To the extent that keyness reflects something otherwise hard to quantify such as aboutness, corpus-importance or text-importance, equal rankings may be deceptive.

Some lists such as those ordered by MI or LL score seem clearly ranked but I suspect a similar restriction might apply there too.


Mike

On 14/10/2016 07:52, Eric Atwell wrote:
Is there a standard metric of overlap between two ranked lists?
e.g. to measure/score the similarity between top 10 keywords extracted
using 2 different formulae, such as LL v MI?
OR e.g. to measure/score the similarity between top 10 hits from Google v top 10 hits from Bing for a give search phrase?
OR e.g. to measure/score the similarity between ranked lists of PoS-tags
predicted for a word by two rival PoS-taggers in an ensemble tagger?

If these were unranked sets of keywords, i could simply count the
intersection. But I want to take rank into account in some senible way.

thanks for expert pointers to proven metrics ...

Eric Atwell, Asst Prof, Language@Leeds and Artificial Intelligence groups, School of Computing, University of Leeds, Times University of the Year 2017

_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora@uib.no
http://mailman.uib.no/listinfo/corpora

--
--
Mike Scott
***
If you publish research which uses WordSmith, do let me know so I can include 
it at 
http://www.lexically.net/wordsmith/corpus_linguistics_links/papers_using_wordsmith.htm
***
Aston University
and
Lexical Analysis Software Ltd
www.lexically.net


_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora@uib.no
http://mailman.uib.no/listinfo/corpora

Reply via email to