1
$\begingroup$

Words like myotonic are biomedical, but words like new appear in regular English texts. I can imagine a few ways to classify words into biomedical and non-biomedical. Has this problem been tackled before? Does anybody know any ready-to-use solutions for this problem?

$\endgroup$
1
  • $\begingroup$ Welcome to the site :) $\endgroup$
    – Dawny33
    Jan 12, 2018 at 16:11

1 Answer 1

1
$\begingroup$

Yes. There has been work going on in the healthcare domain which involves NLP.

Resources like Pubmed have been put together for the exact same purpose.

You can make use of the freely available Pubmed vectors for training your models like maybe a sentence tagging style model for identification. We(my team at work) use them for identification of medical terms like diseases, medicines, etc.

$\endgroup$
4
  • $\begingroup$ I don't need advanced identification of disease names or medicines. I only need to determine if a phrase is biomedical or not. Is there a ready to use tool? Could you show an example on how use bio.nlplab.org to determine whether a phrase is biomedical or not? $\endgroup$ Jan 13, 2018 at 7:19
  • $\begingroup$ @user1424739 You just need to check if the word exists in the Pubmed vector or not. $\endgroup$
    – Dawny33
    Jan 13, 2018 at 9:43
  • $\begingroup$ But a word that appears in Pubmed vector does not mean the word is biomedical. $\endgroup$ Jan 13, 2018 at 14:30
  • $\begingroup$ No. But, you would be close. This is the only research I found related to this topic. $\endgroup$
    – Dawny33
    Jan 13, 2018 at 16:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.