Beefy Boxes and Bandwidth Generously Provided by pair Networks
There's more than one way to do things

Re: counting pairwise incidences

by LanX (Chancellor)
on May 17, 2011 at 14:18 UTC ( #905276=note: print w/replies, xml ) Need Help??

in reply to counting pairwise incidences

If I understand you right, you need to count each pairwise combination in a hash.

To avoid confusion between the combinations ("A","B") and ("B","A") you need to "normalize" i.e. in this case to sort the key "vector" into a unique "representative" (here ("A","B")).

Now you can either use a HoH $h{A}{B}++ or use this old multi-dim feature from Perl4 by separting keys by comma $h{A,B}++!

I prefer the latter.

Code example to follow...

UPDATE: IIRC comma as separator is more secure than using an arbitrary separator like "_" but since I can't find the docs for multi dim keys and your input already uses space as separator, you should stick with it. I.e. changing the code below to $count{"@keys[$a,$b]"}++

{ "aesG gly4" => 2, "aesG phil" => 2, "aesG tomD" => 2, "gly4 phil" => 2, "gly4 tomD" => 3, "phil tomD" => 2, }

Cheers Rolf


use Data::Dump "pp"; my %count; while(<DATA>) { chomp; @keys= sort split; for my $a (0..$#keys){ for my $b ($a+1.. $#keys) { # pp [$keys[$a],$keys[$b]]; $count{$keys[$a],$keys[$b]}++ } } } pp \%count; __DATA__ tomD gly4 phil aesG tomD gly4 phil aesG phil aesG tomD gly4


{ "aesG\34gly4" => 2, "aesG\34phil" => 2, "aesG\34tomD" => 2, "gly4\34phil" => 2, "gly4\34tomD" => 3, "phil\34tomD" => 2, }

2) see $;

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://905276]
and all is quiet...

How do I use this? | Other CB clients
Other Users?
Others wandering the Monastery: (8)
As of 2017-07-24 10:01 GMT
Find Nodes?
    Voting Booth?
    I came, I saw, I ...

    Results (350 votes). Check out past polls.