Bleu Score (Optional) - Sequence Models & Attention Mechanism | Coursera

Course Q&A Deep Learning Specialization Sequence Models
1

I have a doubt regarding the value of the denominator for Precision in the video lecture on Blue Score. Is the denominator equal to the number of words in the machine translated sentence or is it equal to the number of times a particular word ‘the’ in this case appears in the machine translaton? Andrew mentions the former for Precision and the latter for modified Precision.

Thanks
Prejth

2

standard precision simply counts the number of matching n-grams between a machine translation and reference where as modified precision helps one to limit this count by comparing the maximum number of times that n-gram appears in the reference.

so the latter is more focused in machine translation using the same words more compared to the reference.

So prof. Ng statement holds true in both conditions.

1 Like
3

Thank you