The Word Sense Disambiguation for Lagos State: Leveraging a Knowledge Base
ABSTRACT
The disambiguation process relies on words and sense relations present in the database. It involves preprocessing, morphological analysis, A WordNet, and disambiguation components to resolve ambiguity. The preprocessing stage prepares input sentences for morphological analysis by reducing words to their root forms. The WordNet database stores words and their Synsets, along with relations and concepts, to disambiguate polysemous words. The disambiguation component leverages information from the other WSD components.
Two experiments are conducted in this study. The first experiment compares the use of a morphological analyzer with WordNet databases without using them. The accuracy achieved in the experiment is 50.75% and 63.95% respectively. The second experiment explores various window sizes to determine the most suitable size. Based on the results, a window size of three is found to be appropriate.
In conclusion, our Knowledge-based WSD approach presents a valuable method to address word sense disambiguation challenges in Lagos State, showcasing promising results in disambiguating polysemous words and enhancing the understanding of the language’s meaning in different contexts
The Word Sense Disambiguation for Lagos State: Leveraging a Knowledge Base, GET MORE COMPUTER SCIENCE PROJECT TOPICS AND MATERIALS