1.1 --- a/damecco.tex Fri Nov 18 16:18:02 2016 +0100
1.2 +++ b/damecco.tex Fri Nov 18 16:45:03 2016 +0100
1.3 @@ -47,7 +47,7 @@
1.4 %% for the whole article with \linenumbers.
1.5 %% \usepackage{lineno}
1.6
1.7 -\journal{Nuclear Physics B}
1.8 +\journal{Discrete Applied Mathematics}
1.9
1.10 \begin{document}
1.11
1.12 @@ -73,20 +73,63 @@
1.13 %% \address{Address\fnref{label3}}
1.14 %% \fntext[label3]{}
1.15
1.16 -\title{}
1.17 +\title{Improved Algorithms for Matching Biological Graphs}
1.18
1.19 %% use optional labels to link authors explicitly to addresses:
1.20 %% \author[label1,label2]{}
1.21 %% \address[label1]{}
1.22 %% \address[label2]{}
1.23
1.24 -\author{}
1.25 +\author{Alp{\'a}r J{\"u}ttner and P{\'e}ter Madarasi}
1.26
1.27 -\address{}
1.28 +\address{Dept of Operations Research, ELTE}
1.29
1.30 \begin{abstract}
1.31 -%% Text of abstract
1.32 +Subgraph isomorphism is a well-known NP-Complete problem, while its
1.33 +special case, the graph isomorphism problem is one of the few problems
1.34 +in NP neither known to be in P nor NP-Complete. Their appearance in
1.35 +many fields of application such as pattern analysis, computer vision
1.36 +questions and the analysis of chemical and biological systems has
1.37 +fostered the design of various algorithms for handling special graph
1.38 +structures.
1.39
1.40 +The idea of using state space representation and checking some
1.41 +conditions in each state to prune the search tree has made the VF2
1.42 +algorithm one of the state of the art graph matching algorithms for
1.43 +more than a decade. Recently, biological questions of ever increasing
1.44 +importance have required more efficient, specialized algorithms.
1.45 +
1.46 +This paper presents VF2++, a new algorithm based on the original VF2,
1.47 +which runs significantly faster on most test cases and performs
1.48 +especially well on special graph classes stemming from biological
1.49 +questions. VF2++ handles graphs of thousands of nodes in practically
1.50 +near linear time including preprocessing. Not only is it an improved
1.51 +version of VF2, but in fact, it is by far the fastest existing
1.52 +algorithm regarding biological graphs.
1.53 +
1.54 +The reason for VF2++' superiority over VF2 is twofold. Firstly, taking
1.55 +into account the structure and the node labeling of the graph, VF2++
1.56 +determines a state order in which most of the unfruitful branches of
1.57 +the search space can be pruned immediately. Secondly, introducing more
1.58 +efficient - nevertheless still easier to compute - cutting rules
1.59 +reduces the chance of going astray even further.
1.60 +
1.61 +In addition to the usual subgraph isomorphism, specialized versions
1.62 +for induced subgraph isomorphism and for graph isomorphism are
1.63 +presented. VF2++ has gained a runtime improvement of one order of
1.64 +magnitude respecting induced subgraph isomorphism and a better
1.65 +asymptotical behaviour in the case of graph isomorphism problem.
1.66 +
1.67 +After having provided the description of VF2++, in order to evaluate
1.68 +its effectiveness, an extensive comparison to the contemporary other
1.69 +algorithms is shown, using a wide range of inputs, including both real
1.70 +life biological and chemical datasets and standard randomly generated
1.71 +graph series.
1.72 +
1.73 +The work was motivated and sponsored by QuantumBio Inc., and all the
1.74 +developed algorithms are available as the part of the open source
1.75 +LEMON graph and network optimization library
1.76 +(http://lemon.cs.elte.hu).
1.77 \end{abstract}
1.78
1.79 \begin{keyword}