DM: RE: CSI DVW Preservation of "Nearness"

[Beck, Hal]

Warren suggests that 
        "Since the original announcement was posted to
> datamine-l@nautilus-sys.com, it would be nice to have an explanation
> posted to that list of what the marketese really means."
I, for one, would appreciate this. We all have customers who depend 
on our thorough understanding of the tools we use; otherwise, how can 
our customers, or ourselves, have confidence in our interpretation of 
the analytic results. Thanks Warren.

Hal Beck

> ----------
> From:         Warren Sarle[SMTP:saswss@unx.sas.com]
> Sent:         Wednesday, January 07, 1998 2:43 PM
> To:   mhancock@csihq.com
> Cc:   datamine-l@nautilus-sys.com
> Subject:      DM: CSI DVW Preservation of "Nearness"
> 
> Monte Hancock <mhancock@csihq.com> writes:
> > Because the DVW regression is continuous and regular, points which
> > are proximate in Rn remain proximate on the display, and in a way
> > which preserves intuitive visual elements of the local geometry.
> 
> In any linear 2D projection, points which are proximate in Rn remain
> proximate on the display. What's special about what CSI Visualizer
> does?
>  
> > The DVW regression preserves many of the same intuitive aspects of
> > proximity which, for example, make a flat photograph of a
> > 3-dimensional scene useful.  Objects are recognizable and
> > interpretable.  High-dimensional theoretical geometric effects can
> > be visually observed (e.g., in 4-space, 3-spheres appear
> > ellipsoidal/flat from certain perspectives, while 4-spheres are
> > "round" from all perspectives).
> 
> These examples are not challenging. What happens with nontrivial
> examples such as those in Furnas & Buja (1994), "Prosection views:
> dimensional inference through sections and projections, J. of 
> Computational and Graphical Statistics, 3, 323-385?
> 
> Since the original announcement was posted to
> datamine-l@nautilus-sys.com, it would be nice to have an explanation
> posted to that list of what the marketese really means.
> 
> -- 
> 
> Warren S. Sarle       SAS Institute Inc.   The opinions expressed 
>here
> saswss@unx.sas.com    SAS Campus Drive     are mine and not
> necessarily
> (919) 677-8000        Cary, NC 27513, USA  those of SAS Institute.
> * Do not send me unsolicited commercial, political, or religious 
>email
> *
>