Word Sense Disambiguation Using Differential Evolution Algorithm

Abstract

Word sense disambiguation (WSD) can be defined as the process of determining the meaning of a word as it is used in a specific context from a set of possible word senses. This task is of significant importance in many applications, such as text summarization, information retrieval, and information extraction. Many researchers have focused on the use of metaheuristic algorithms to find the most suitable sense that reflects the meaning of ambiguous words. However, the use of metaheuristic approaches remains limited and therefore requires efficient exploration of the problem space. In this paper, a new method is proposed to address the WSD problem based on differential evolution. The proposed method consists of two main phases. The first phase concerns estimating a numerical value for word relatedness based on a shortest-path measure. In the second phase, a differential evolution algorithm is applied to select the best sense corresponding to the ambiguous word. The proposed method was evaluated using the SemCor dataset, and performance was measured using three evaluation metrics: recall, precision, and F-score. The results demonstrate that the proposed method outperforms related approaches, particularly in terms of F-score and precision, achieving values of 85.6% and 89.2%, respectively.