Using Table based Version of K Nearest Neighbor for Classifying Words Semantically

This article proposes the modified KNN (K Nearest Neighbor) algorithm which receives a table as its input data and is applied to the word categorization. The motivations of this research are the successful results from applying the table based algorithms to the text categorizations in previous works and the expectation of synergy effect between the text categorization and the word categorization. In this research, we define the similarity metric between two tables representing words, modify the KNN algorithm by replacing the exiting similarity metric by the proposed one, and apply it to the word categorization. The proposed KNN is empiricallyvalidated as the better approach in categorizing words in news articles and opinions. In using the table based KNN algorithm, it is easier to trace results from categorizing words.