Part-of-speech tagging using evolutionary computation

Abstract

Part-of-speech tagging is a task of considerable importance in the field of natural language processing. Its purpose is to automatically tag the words of a text with labels that designate the appropriate parts-of-speech. The approach proposed in this paper divides the problem into two tasks: a learning task and an optimization task. Algorithms from the field of evolutionary computation were adopted to tackle each of those tasks. We emphasize the use of swarm intelligence, not only for the good results achieved, but also because it is one of the first applications of such algorithms to this problem. This approach was designed with the aim of being easily extended to other natural language processing tasks that share characteristics with the part-of-speech tagging problem. The results obtained in two different English corpora are among the best published.