The example from the paper doesn't provide the same result as in the paper

[charlie42]

In the paper the sentences 'A gem is a jewel or stone that is used in jewellery.' and 'A jewel is a precious stone used to decorate valuable things that you wear, such as rings or necklaces.' result in dot 3.31974454153 and similarity 0.8534.

When I run the script on the same sentences:

a = 'A gem is a jewel or stone that is used in jewellery.'
b = 'A jewel is a precious stone used to decorate valuable things that you wear, such as rings or necklaces.'

I get these results:

dot 2.83492612731367
similarity 0.566985225462734

[nmtoan91]

faked result