文摘
Word sense disambiguation is an early problem in the field of computational linguistics, and is defined as identifying the sense (or senses) that most likely represents a word, or a sequence of words in a given context. Word sense disambiguation was recently addressed as a combinatorial optimization problem in which the goal is to find a sequence of senses that maximizes the semantic relatedness among the target words. In this paper, we propose a novel algorithm for solving the word sense disambiguation problem, namely D-Bees, that is inspired by the bee colony optimization meta-heuristic in which several artificial bee agents collaborate to solve the problem. The D-Bees algorithm is evaluated on a standard SemEval 2007 task 7 coarse-grained English all-words corpus and is compared to the genetic and simulated annealing algorithms as well as an ant colony algorithm. It will follow that the bee and ant colony optimization approaches perform on par achieving better results than the genetic and simulated annealing algorithms on the given dataset.