Problem statement: This study attempts to present an object-net method for word sense disambiguation. It is proposed to model the elementary meanings which assist the machine to autonomously undertake the analysis and synthesis processes of meaning. Approach: In the proposed methodology, the disambiguation process was performed in context manner. Starting from natural text, the context of the sentence was identified, then the actual meaning identified using correlation of elementary object meanings existed in object-net database. It was because even ambiguous word will have only one meaning based on the context or object or domain on which the sentence was written. Results: This object-net approach disambiguates original text with high precision of 96% of the verbs and 97% of nouns for data extraction from the database and reporting in terms of graphs. Conclusion: The accuracy of finding the sense of a word and extracting data from the database and projecting into graphs was based on number of trained objects in object-net database. Due to this object-net database plays a major role in this proposed method.
