TFIDF : Data Science Concepts | Tin Hay về chủ đề tf idf |

This is really good. Concise , straight to the point, and there is no need to show a line of code !

So clear your explanation! Thanks a lot!!:)))

This is a great explanation. Thanks.
I have a question about differences between the implementation described in this video and another implementation commonly found on the web.
Can you explain how these two details would impact the final representation:
1) Term frequency simply calculated as term count
2) Applying vector normalisation (L2) to the document vector obtained in this video

Another question which is more open-ended: why is TfIdf still relevant ? Or less provocatively – is there a sweet spot where one would prefer TfIdf over the modern dense vector representations (such as word2vec, doc2vec, etc.) ?

Being a math lover, within a minute of your explanation I became your fan, was always in a search of videos like this

Excellent teaching! Perfectly designed, clearly explained and not even one sentence that would be redundant. I’m your fan my friend 👍🏼🙏🏼

Awesome video!!

I like your videos first and then start watching your Data Science videos because I am sure that after I am done watching it, I will like it anyway.
Keep it up.. 🙏

Good explanation in a simple way… keep doing well man

How do you model multiple objects associated to a term class: Dental Care: United Health Care, Blue Shield, …, by state? This becomes contextual and local within the text – how close is the word dental care in the text to UHC, for instance. The result would show which states address dental care in their health insurance regulations and which insurance companies make it available – both in a positive and negative way. Understand that this is a narrow example. Thanks

I read this explanation in a book, but not as clear as this video. Well done!