It would be great if wordsegment returned that.
For instance:
> ws.rank("nobodyelse")
[ ["nobody", "else"], ["no", "body", "else"], ...]
or
> ws.probabilities("nobodyelse")
[
[ ["nobody", "else"], ["no", "body", "else"], ...],
[ 0.727362, 0.0012372, ...]
]