[R] Evidence Theory in R

[H. W. Borchers]

Lars Fischer <lars <at> sec.informatik.tu-darmstadt.de> writes:

> Hello,
> 
> well, I searched list-archive, cran and the references, but found
> nothing. Thus:
> 
> Does anybody around here know anything about Dempster-Shafer Theory,
> Evidence Theory or Hints in R? Has anybody stumbled about a package that
> I overlooked or implemented something in this area? I really would like
> to not implement a hint-model a second time.
----

Dempster-Shafer Theory was (almost) a hype 15 or more years ago when 
it was one of several reasoning systems. It has been heavily discussed 
and criticized, and it was considered by many as statistically unsound. 

This might be the reason why it has not been realized in an R package.

As a scheme for reasoning under uncertainty Dempster-Shafer Theory has 
been replaced by Bayesian Networks and their learning capabilities in 
many modern knowledge-based systems.

Fortunately, there are several packages in R that implement Bayesian 
statistics and networks, including learning facilities. You will find 
references in the 'Bayesian' and 'Machine Learning' task views.

Regards,  Hans Werner

----
> My apologies if I missed something obvious, but I would be glad if
> someone could give me a hint.
> 
> Regards,
> 	Lars