AbstractWord sense disambiguation is a subfield of computational linguistics in which computer systems are designed to determine the appropriate meaning of a word as it appears in the linguistic context. This article provides a survey of what has been done in this area: the ways that word meaning can be represented in the computer, the approaches taken by systems, how performance is evaluated and an overview of the intended applications that might benefit from this technology. One of the major issues has been, and still remains, that of finding an appropriate computational representation of word meaning as this is fundamental to the performance and utility of systems.