about ranges in MDS

[Loet Leydesdorff]

Dear Jose Luis,

You get more spreading with the Pearson correlation.

See: Jones, W. P., & G. W. Furnas. (1987). Pictures of Relevance: A
Geometric Analysis of Similarity Measures. Journal of the American Society
for Information Science, 36(6), 420-442.

With kind regards,


Loet

> -----Original Message-----
> From: ASIS&T Special Interest Group on Metrics
> [mailto:SIGMETRICS at LISTSERV.UTK.EDU] On Behalf Of José Luis Ortega
> Sent: Friday, January 21, 2005 1:46 PM
> To: SIGMETRICS at LISTSERV.UTK.EDU
> Subject: [SIGMETRICS] about ranges in MDS
>
> I have a question for mathematicians:
>
> The cosine have a range [0,1] and Pearson's Coefficient
> [1,-1]. If i build a matrix distance with these measures, can
> theses different ranges affect in a MDS display of the data?
> and How? more spreading or less?
>
> Thank you
>
> José Luis Ortega Priego
> Internet Lab
> CINDOC-CSIC
> Joaquín Costa, 22
> 28002 Madrid(SPAIN)
> http://internetlab.cindoc.csic.es/miembros.asp?id=11
>