1.1 --- a/doc/groups.dox	Thu Sep 13 12:23:46 2012 +0200
     1.2 +++ b/doc/groups.dox	Wed Nov 07 17:39:39 2012 +0100
     1.3 @@ -407,9 +407,19 @@
     1.4     strongly polynomial \ref klein67primal, \ref goldberg89cyclecanceling.
     1.5  
     1.6  In general, \ref NetworkSimplex and \ref CostScaling are the most efficient
     1.7 -implementations, but the other two algorithms could be faster in special cases.
     1.8 +implementations.
     1.9 +\ref NetworkSimplex is usually the fastest on relatively small graphs (up to
    1.10 +several thousands of nodes) and on dense graphs, while \ref CostScaling is
    1.11 +typically more efficient on large graphs (e.g. hundreds of thousands of
    1.12 +nodes or above), especially if they are sparse.
    1.13 +However, other algorithms could be faster in special cases.
    1.14  For example, if the total supply and/or capacities are rather small,
    1.15  \ref CapacityScaling is usually the fastest algorithm (without effective scaling).
    1.16 +
    1.17 +These classes are intended to be used with integer-valued input data
    1.18 +(capacities, supply values, and costs), except for \ref CapacityScaling,
    1.19 +which is capable of handling real-valued arc costs (other numerical
    1.20 +data are required to be integer).
    1.21  */
    1.22  
    1.23  /**
    1.24 @@ -448,7 +458,7 @@
    1.25  \brief Algorithms for finding minimum mean cycles.
    1.26  
    1.27  This group contains the algorithms for finding minimum mean cycles
    1.28 -\ref clrs01algorithms, \ref amo93networkflows.
    1.29 +\ref amo93networkflows, \ref karp78characterization.
    1.30  
    1.31  The \e minimum \e mean \e cycle \e problem is to find a directed cycle
    1.32  of minimum mean length (cost) in a digraph.
    1.33 @@ -464,12 +474,11 @@
    1.34  function.
    1.35  
    1.36  LEMON contains three algorithms for solving the minimum mean cycle problem:
    1.37 -- \ref KarpMmc Karp's original algorithm \ref amo93networkflows,
    1.38 -  \ref dasdan98minmeancycle.
    1.39 +- \ref KarpMmc Karp's original algorithm \ref karp78characterization.
    1.40  - \ref HartmannOrlinMmc Hartmann-Orlin's algorithm, which is an improved
    1.41 -  version of Karp's algorithm \ref dasdan98minmeancycle.
    1.42 +  version of Karp's algorithm \ref hartmann93finding.
    1.43  - \ref HowardMmc Howard's policy iteration algorithm
    1.44 -  \ref dasdan98minmeancycle.
    1.45 +  \ref dasdan98minmeancycle, \ref dasdan04experimental.
    1.46  
    1.47  In practice, the \ref HowardMmc "Howard" algorithm turned out to be by far the
    1.48  most efficient one, though the best known theoretical bound on its running