Comment
Author: Admin | 2025-04-27
As: R l o g F ( p a t t e r n ( i ) ) = ( F i N i ) ∗ log 2 F i , where F i represents the number of category members extracted by p a t t e r n ( i ) and N i is the total number of nouns extracted by p a t t e r n i . This formula was used to score the patterns with a high precision or moderate precision but a high recall. The high scoring patterns were then placed in the pattern pool. After this process, all head nouns co-occurring with patterns in pattern pool were added to the candidate word pool. At the end of each bootstrapping cycle, the best candidates were added to the lexicon thus enlarging the lexicon set. 2.2.2. Identify Stressed or Non-Stressed Tweets Using Words Obtained from Basilisk AlgorithmThe process used related to Basilisk, as proposed by Thelen and Riloff, can be described using the algorithm shown on Table 3 (for notation description see Appendix A). This performs the categorization task of assigning nouns in an unannotated corpus to their corresponding semantic categories. Using the words generated by the Basilisk algorithm, we counted the total number of occurrences of any of the keywords in both categories. After the total count of stress and non-stress words in each tweet was obtained, we determined whether the tweet was in the category of stressed or non-stressed or neutral. This
Add Comment