NLP Tutorial in Python - Spam Classification

In this video we implement an email spam classifier using NLTK (natural language processing toolkit) in Python. We use the bag of words (BOW) approach to building the model, after performing tokenizing, lemmatization / stemming, and removing stop words. You'll learn a ton about NLP in just 20 minutes!
Link to the Colab notebook: colab.research...
Thank you for watching the video! You can learn data science FASTER at mlnow.ai!
Master Python at mlnow.ai/cours...!
Learn SQL & Relational Databases at mlnow.ai/cours...!
Learn NumPy, Pandas, and Python for Data Science at mlnow.ai/cours...!
Become a Machine Learning Expert at mlnow.ai/cours...!
Don't forget to subscribe if you enjoyed the video :D

[@GregHogg]

Take my courses at mlnow.ai/!

[@duchahapsari7081]

Wow! This is an amazing tutorial. Learn so much about fundamental NLP, while keeping it so practical to follow. Looking for more crisp contents from you, Greg!

[@GregHogg]

Really glad to hear that Ducha!

[@ashathotan]

I enjoyed watching your illustration on the email spam.

[@DarkTobias7]

Amazing video, please do more NLP projects

[@GregHogg]

Thanks so much, and will do!

[@Brocollipy]

Great video , just what I needed. Need to test out some models quickly and really can't do another 3hr course!!

[@arsheyajain7055]

This is awesome 👏

[@GregHogg]

I thought so too!

[@gustavojuantorena]

Great video!

[@GregHogg]

Thanks so much Gustavo!

[@akshitadixit_1068]

Thanks Greg for this amazing video, however i have a doubt, how are we determining that the tokens with the maximum frequency are the ones contributing to the spamy nature of the message, their is a possibility that tokens with a low frequency are equally malicious.
Thanks!

[@GregHogg]

Yes that's very true, you could absolutely change how I did things

[@jspetrolina]

Hi Greg! Old but gold, fabulous video, let me ask you, is that structure that you build what packages like spacy is doing behind the scenes? Thanks again

[@GregHogg]

Thank you :) and what structure sorry?

[@iqrarkhan8129]

thanks that was quite helpful bus can you also please do a malware detection and classification using machine learning algo? if yes please do and upload as soon as possible

[@GregHogg]

You're very welcome! Probably eventually, but won't be able to do that for awhile sorry.

[@mikekertser5384]

Thank you! Can you please make a video with the review of the word embeddings models and corresponding transfer learning examples?
And some nlp feature engineering as well. :)

[@GregHogg]

You're very welcome - and this is in the works :)

[@InfernalPasquale]

12:00 Why is features = set() converted to a list, rather than just being a list to begin with?

[@rahulnayak8866]

using a list instead of set will result in getting redundant values, so in order to get unique values set() is used which is then converted into a list.

[@e_hossam96]

This is great

[@GregHogg]

Thanks so much Hossam!! :)

[@writabratadey8048]

nltk.download() is not working. showing WinError 10060 everytime ie. connection attempt failed...plz provide a solution

[@GregHogg]

Probably slow internet unfortunately

[@MyStockz]

Hi Greg Hogg! Hope all is well! Have you tried or heard of a website called logikbot? If yes, what do you think of it?

[@GregHogg]

Never heard :)

[@MyStockz]

@@GregHogg Cool!

[@tareq8109]

Want NLP series

[@GregHogg]

Working on it!

[@dannyrodin1151]

I enjoy your videos, but this one's way too fast. I'll need to watch it 3 times with 0.5 speed.

[@GregHogg]

Hmm, I appreciate the feedback here.