[Corpora-List] tf-idf vs. cosine in natural texts

Chris Jordan chris.jordan at acm.org
Wed Dec 17 11:07:46 UTC 2008


TF-IDF is used for assigning weight to terms in a document.
http://en.wikipedia.org/wiki/Tfidf

Cosine similarity is used for comparing term vectors that represent  
documents.
http://en.wikipedia.org/wiki/Cosine_similarity

On 17-Dec-08, at 2:10 AM, KHALED ABDALGADER wrote:

> Hi
>
>
> I'm curious: what's the difference between using TF-IDF and cosine  
> as the retrieval model in natural texts?
>
>
>
> Perhaps someone can point me to the difference or offer a brief  
> description. Please ...
>
>
>
> Thanks in advance
>
>
>
> Khaled
>
>
> _______________________________________________
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora


_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora



More information about the Corpora mailing list