IOB Accuracy in named entity classifiers

[Rafael Parpinel Cavina]

Hi

I'm doing some works on NLTK with named entity recognition and chunkers. I retrained a classifier usingnltk/chunk/named_entity.py for that and I got the following mesures:

ChunkParse score:
    IOB Accuracy:  96.5%
    Precision:     78.0%
    Recall:        91.9%
    F-Measure:     84.4%

But I don't understand what is the exact difference between IOB Accuracy and Precision in this case. Actually, I found on the docs (here) the following for an specific example:

The IOB tag accuracy indicates that more than a third of the words are tagged with O, i.e. not in an NP chunk. However, since our tagger did not find any chunks, its precision, recall, and f-measure are all zero.

So, if IOB accuracy is just the number of O labels, how come we don't have chunks and IOB accuracy is not 100% at the same time, in that example?

Thank you in advance