It needs to be rule based. On Thu, Mar 5, 2020, 12:06 Himanshu choudhary < himanshuchoudhary_bt2...@dtu.ac.in> wrote:
> Hi, > > I just want to ask can't we use Neural machine translation or unsupervised > machine translation rather than rule-based learning for the task "Apertium > English--Hausa/Igbo/Swahili/Tigrinya/Yoruba". As I got some open-source > data for some of these languages and I believe neural machine translation > can produce better results and also we don't need to have a knowledge of > the morphology and grammar of that particular language as it is very rare > that some developer has the knowledge of these languages. Please let me > know if I can proceed with it or I have to apply rule-based learning. so > that I can move forward. > > Thank you > > _______________________________________________ > Apertium-stuff mailing list > Apertium-stuff@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/apertium-stuff >
_______________________________________________ Apertium-stuff mailing list Apertium-stuff@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/apertium-stuff
