On Word Embeddings

Word embeddings are more suitable for synonyms. It gives decent result for synonyms.  Word classification can get a bit tricky with word embeddings. Maybe some modifications will make it useful. I think training on domain specific data will give better results. Anyone has pretrained embeddings on medical domain?

I will launch a word similarity api from different sources.

Comments

Popular posts from this blog

On Quora

No Marriage

Pls Don't Fuck My Leg