Building a robust sentiment lexicon with (almost) no resource

12/15/2016

∙

by   Mickael Rouvier, et al.

∙

0

∙

share

Creating sentiment polarity lexicons is labor intensive. Automatically translating them from resourceful languages requires in-domain machine translation systems, which rely on large quantities of bi-texts. In this paper, we propose to replace machine translation by transferring words from the lexicon through word embeddings aligned across languages with a simple linear transform. The approach leads to no degradation, compared to machine translation, when tested on sentiment polarity classification on tweets from four languages.