A modified artificial bee colony algorithm for solving large graph theory problems

The artificial bee colony (ABC) algorithm has been successfully used to find good solutions to real parameter optimization problems. Unfortunately, solutions to graph theory problems are often encoded as binary strings. In this paper we propose a modified ABC problem suitable for finding solutions to large graph theory problems. The ABC algorithm searches in a large dimension unit hypercube and the real-valued components are converted into binary substrings, which can be concatenated to form a potential solution to a graph theory problem. A different method of deploying scout bees is also incorporated to help escape local optima. Our modified ABC algorithm is tested on two difficult large graph problems.