Training Spacy's Named Entity Recognition to Recognize Drugs – NLP in Python


In this tutorial we will learn how to create a dataset and train Spacy’s Named Entity Recognition to identify Drugs as a new entity using the Drug Reviews Dataset.

Udemy Course : Building ML Web Apps

Official Docs:
Credits :@curiousprogrammer

If you liked the video don’t forget to leave a like or subscribe.
If you need any help just message me in the comments, you never know it might help someone else too.
J-Secur1ty JCharisTech
Jesus Saves

==Get The Data Science Prime App==
@ Playstore :

==Need To Build A Data Science/ML App Check out this gig==




Comment List

  • Lead Learner
    January 22, 2021

    Cant find the notebook in the github link!

  • Lead Learner
    January 22, 2021

    Is the TEXT body important for TEXT , annotation in TRAIN_DATA ? 

    Its a lot simpler to just train the NER directly off the name of the drug rather than having to manually/automatically create an entire sentence around the drug name..say for example:

    medList = ['ibuprofen' , ''paracetemol' , 'panadol' ]
    TRAIN_DATA = [(med_name , { 'entities' : [ (0 , len(med_name)-1 , 'DRUG' ) ] } ) for med_name in medList ]

    You reckon any issues/downside with training the NER with just the drug name only?

    Thanks for the tutorials nonetheless!

  • Lead Learner
    January 22, 2021

    excellent tutorial, why we are doing count < 1000 here?

  • Lead Learner
    January 22, 2021

    Thank you for this amazing tutorial. I have a question though, What if there is no column such as drug_name in the dataset?. How do we train the model in that case?

  • Lead Learner
    January 22, 2021

    Thank you very much it is a great tutorial, can please upload tutorial on pyTorch for text

Write a comment