Table 2 Mean contribution of algorithms to word list generation
From: A UMLS-based spell checker for natural language processing in vaccine safety
WORD LIST ALGORITHM | TRAINING SET n = 12,056 | TEST SET n = 8,131 |
|---|---|---|
N-Gram | 20% | 13% |
Header | 55% | 59% |
Metaphone | 8% | 4% |
Transposition | 1% | 3% |
Deletion | 5% | 6% |
Substitution | 5% | 5% |
Insertion | 6% | 10% |
TOTAL | 100% | 100% |