That answer on Spanish SE is misleading on key points - "neural networks" have nothing to do with dictionaries. Let's step back and imagine that we are tasked with creating bilingual dictionaries for many language pairs. To start, we have human-compiled ones, either from processing Wiktionary entries' Translations sections, or purchased from companies ...


The BLEU score is between 0 and 1, but is sometime expressed as a percentage, i.e. ranging from 0 to 100%. E.g. http://www.statmt.org/moses/?n=Moses.SupportTools#ntoc5 returns a score between 0 and 100 (code). Misc: Original BLEU paper: http://www.aclweb.org/anthology/P02-1040.pdf Some technical issues in BLEU: https://github.com/nltk/nltk/issues/1268


Hope it is not too late. Both ppl and ppl1 are normalized according http://www.speech.sri.com/projects/srilm/manpages/srilm-faq.7.html , which is called entropy rate.


Let me post a slightly tongue-in-cheek answer, then perhaps expand on it in useful directions. If your corpus is 2.3 million words, the ideal gold standard is 2.3 million words. Now, to try to frame that, what I really mean is that you are approaching this from the wrong direction. You are not creating a corpus to satisfy your statistics teacher; you are ...


The recall metric is ignored when evaluating syntactic trees because all tokens are being labeled in one way or another. There are 5 most common metrics for the evaluation of syntactic dependency parsing: Head Attachment Score (percent of nodes which are correctly attached to their parent) Label Precision (percent of nodes whose dependency labeled is ...

Only top voted, non community-wiki answers of a minimum length are eligible