I am working on sentiment analysis in Map Reduce (Java) using the Stanford Core NLP API. I have integrated them successfully and have noticed in the console that it specifically works on all the input lines. It takes a lot of time in my machine( 3GB RAM). So will it same same proportion of time in other powerful machines or there is a solution for reducing the time ?
Anybody having idea of this running time in other machines can share their experience.
Will anybody suggest anything about Amazon Web Services( AWS ) for this?
I am also trying to run corenlp on aws. Have you fixed your problem? Which Machine suits for this purpose? Can you help on this topic?
RépondreSupprimer