[R] Compare two data sets

[jim holtman]

Easiest way to do it is to try it out and time it.  Here is a case
where I generated two sets of data with 120,000 characters each (just
random numbers converted to character strings) and then asked for the
intersection of them.  Came up with 3 matched in about 0.2 seconds.
That would seem fastest enough, unless you plan to do this operation
tens of thousands of times:

> x <- as.character(runif(120000))
> y <- as.character(runif(120000))
> system.time(z <- intersect(x,y))
   user  system elapsed
   0.22    0.00    0.22
> str(z)
 chr [1:3] "0.289942682255059" "0.75132836541161" "0.638638160191476"
>

Here is the timing if you get 50000 matches and it is about the same:

> x <- as.character(round(runif(120000),5))
> y <- as.character(round(runif(120000),5))
> system.time(z <- intersect(x,y))
   user  system elapsed
    0.2     0.0     0.2
> str(z)
 chr [1:48908] "0.08385" "0.62639" "0.47603" "0.18578" "0.89447"
"0.58435" "0.15297" ...
>


On Tue, Mar 25, 2008 at 10:28 PM, Suhaila Zainudin
<suhaila.zainudin at gmail.com> wrote:
> Hi,
>
> Thanks for the feedback. I have tried it on the small size sample and ref
> and it works. Now I want to use a larger dataset for myref (the reference
> file) . The reference file contains 112189 rows. Can I use the same approach
> that works for the small example? Or are there other alternatives when
> dealing with data of that magnitude?
>
>
> --
> Suhaila Zainudin
> PhD Candidate
> Universiti Teknologi Malaysia



-- 
Jim Holtman
Cincinnati, OH
+1 513 646 9390

What is the problem you are trying to solve?