######################################################################
   Algorithm::SetCovering 0.05
######################################################################

NAME
   Algorithm::SetCovering - Algorithms to solve the "set covering problem"

SYNOPSIS
       use Algorithm::SetCovering;

       my $alg = Algorithm::SetCovering->new(
           columns => 4,
           mode    => "greedy");

       $alg->add_row(1, 0, 1, 0);
       $alg->add_row(1, 1, 0, 0);
       $alg->add_row(1, 1, 1, 0);
       $alg->add_row(0, 1, 0, 1);
       $alg->add_row(0, 0, 1, 1);

       my @to_be_opened = (@ARGV || (1, 1, 1, 1));

       my @set = $alg->min_row_set(@to_be_opened);

       print "To open (@to_be_opened), we need ",
             scalar @set, " keys:\n";

       for(@set) {
           print "$_: ", join('-', $alg->row($_)), "\n";
       }

DESCRIPTION
   Consider having M keys and N locks. Every key opens one or more locks:

            | lock1 lock2 lock3 lock4
       -----+------------------------
       key1 |   x           x
       key2 |   x     x
       key3 |   x     x     x
       key4 |         x           x
       key5 |               x     x

   Given an arbitrary set of locks you have to open (e.g. 2,3,4), the task
   is to find a minimal set of keys to accomplish this. In the example
   above, the set [key4, key5] fulfils that condition.

   The underlying problem is called "set covering problem" and the
   corresponding decision problem is NP-complete.

 Methods
   $alg = Algorithm::SetCovering->new(columns => $cols, [mode => $mode]);
       Create a new Algorithm::SetCovering object. The mandatory parameter
       "columns" needs to be set to the number of columns in the matrix
       (the number of locks in the introductory example).

       "mode" is optional and selects an algorithm for finding the
       solution. The following values for "mode" are implemented:

       "brute_force"
           Will iterate over all permutations of keys. Only recommended for
           very small numbers of keys.

       "greedy"
           Greedy algorithm. Scales O(mn^2). Can't do much better for a
           NP-hard problem.

       The default for "mode" is set to "greedy".

   $alg->add_row(@columns)
       Add a new row to the matrix. In the example above, this adds one key
       and specifies which locks it is able to open.

           $alg->add_row(1,0,0,1);

       specifies that the new key can open locks #1 and #4.

       The number of elements in @columns needs to match the previously
       defined number of columns.

   $alg->min_row_set(@columns_to_cover)
       Determines a minimal set of keys to cover a given set of locks and
       returns an array of index numbers for those keys.

       Defines which columns have to be covered by passing in an array with
       true values on element positions that need to be covered. For
       example,

           my @idx_set = $alg->min_row_set(1,1,0,1);

       specifies that all but the third column have to be covered and
       returns an array of index numbers into an array, defined previously
       (and implicitely) via successive add_row() commands.

       If no set of keys can be found that satisfies the given requirement,
       an empty list is returned.

       If you've forgotten which locks the key referred to by a certain
       index number can open, use the "rows()" method to find out:

           my(@opens_locks) = $alg->rows($idx_set[0]);

       will give back an array of 0's and 1's, basically returning the very
       parameters we've passed on to the add_row() command previously.

 Strategies
   Currently, the module implements the Greedy algorithm and also (just for
   scientific purposes) a dumb brute force method, creating all possible
   combinations of keys, sorting them by the number of keys used
   (combinations with fewer keys have priority) and trying for each of them
   if it fits the requirement of opening a given number of locks.

   This obviously won't scale beyond a really small number of keys (N),
   because the number of permutations will be 2**N-2.

   The Greedy Algorithm, on the other hand scales with O(mn^2), with m
   being the number of keys and n being the number of locks.

 Limitations
   Julien Gervais-Bird <[email protected]> points out: The greedy
   algorithm does not always return the minimal set of keys. Consider this
   example:

            | lock1 lock2 lock3 lock4 lock5 lock6
       -----+------------------------------------
       key1 |   x           x           x
       key2 |   x     x
       key3 |               x     x
       key4 |                           x     x

   The minimal set of keys to open all the locks is (key2, key3, key4),
   however the greedy algorithm will return (key1,key2,key3,key4) because
   key1 opens more locks than any other key.

AUTHOR
   Mike Schilli, 2003, <[email protected]>

   Thanks to the friendly guys on rec.puzzles, who provided me with
   valuable input to analyze the problem and explained the algorithm:

       Craig <[email protected]>
       Robert Israel <[email protected]>
       Patrick Hamlyn <[email protected]_ocomsp_am.au>

COPYRIGHT AND LICENSE
   Copyright 2003 by Mike Schilli

   This library is free software; you can redistribute it and/or modify it
   under the same terms as Perl itself.

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.