Table 5.
Degree of overlap on tokens and part-of-speech tags
Tokens | Part-of-speech tags | |||||||
---|---|---|---|---|---|---|---|---|
Feature request | ReadMe | JavaDoc | Avg | Feature Request | ReadMe | JavaDoc | Avg | |
NLTK vs. spaCy | 92% | 95% | 97% | 95% | 79% | 84% | 85% | 83% |
NLTK vs. CoreNLP | 95% | 88% | 99% | 94% | 75% | 71% | 82% | 76% |
NLTK vs. OpenNLP | 93% | 94% | 98% | 95% | 82% | 84% | 88% | 85% |
spaCy vs. CoreNLP | 89% | 94% | 97% | 94% | 74% | 79% | 83% | 79% |
spaCy vs. OpenNLP | 89% | 94% | 96% | 93% | 80% | 85% | 86% | 84% |
CoreN vs. OpenNLP | 90% | 87% | 98% | 92% | 77% | 75% | 84% | 79% |