I am Bengali as well. I am glad that you are doing this and feeling proud of you brother.

Rocked in very less time. This can only come with very keen knowledge.

excellent vids brother ... keep going. You've compressed months & months of learning into a easy-to-learn videos ... please don't stop

@NormalizedNerd

4 жыл бұрын

Thanks a lot for appreciating the effort :D

thanks a lot man, after two days of such boring videos on youtube finally found a good one to learn NLP, great work man keep going on machine learning and other data science topic, it's really rare to find such a great video.

@NormalizedNerd

3 жыл бұрын

Thanks a lot mate! Keep supporting...

I like the video. Simple and straight to the point. Keep it up!

@NormalizedNerd

3 жыл бұрын

Thanks! :D

Thank you for your best explanation!

Amazing. W8ing for more video😊

Really Amazing and very clear . Keep this up . New subscriber .

@NormalizedNerd

4 жыл бұрын

Thanks for the sub!

excellent. thanks Normalised Nerd

My 4 months struggle to learn nlp got done in 23 min. Thanks a ton bro!

@NormalizedNerd

3 жыл бұрын

Glad to hear that bro! ❤

Very nice explanation also covered much things in very less time. Keep it up Man👍

@NormalizedNerd

3 жыл бұрын

Glad to hear that!

I am at present studying the Machine Learning A-Z course from Udemy. Trust me, the tutor has no Idea how to explain stuff and it is the best selling course on Udemy for Data Science. I scratched my head while I was going through his tutorial, then I came here and all my doubts were cleared. Coding in python is not a difficult task but understanding the concept is the most important thing. And I got that understanding of Bag of words Model from your tutorial. Thanks a lot for your help. P.S. One who understands the concept and has strong Fundamentals has the ability to explain stuff in the simplest manner possible. Please keep it Simple like this in upcoming tutorials also. ALREADY SUBSCRIBED.

@NormalizedNerd

3 жыл бұрын

Thanks a lot @Sumit Chhabra. I'll try my best to maintain this level of simplicity :)

Thank you so much

Thank you sir 🙏

Thanks brother❤

Well explained !!!

@NormalizedNerd

3 жыл бұрын

Thanks!!

you are great.

Very informative video , thank you for uploading the NLP series , having one queries how can we use auto text summary generator in other human lanaguages text like(Japanese , chinese & Korean)????? your reply would be very helpfull.

@NormalizedNerd

3 жыл бұрын

Unfortunately there's no library for that can summarize every language. However, you'll find many github repos where people have built text summarizers for other languages using the same method!

Please make a video or two about neural machine translation. With an example.

amazing keep posting videos :)

@NormalizedNerd

4 жыл бұрын

Will do 😁

chalie jao dada!!!

@NormalizedNerd

4 жыл бұрын

Thank you bhai!

@thetensordude

4 жыл бұрын

Can you make some videos about seaborn?

Shouldn't you initialize the regular expression outside the loop?

What does CountVectorizer do in this model ? Does it convert the words in the instances/Document into 0/1 please suggest.... Thanks for your help in advance.

@NormalizedNerd

3 жыл бұрын

It'll generate the feature matrix that I started drawing at 7:06

Great, New subscriber

@NormalizedNerd

4 жыл бұрын

Thanks a lot :D

Hello Normalize Nerd, i've got an error in y(target) y = data.as_matrix(['Review_class'], AttributeError: 'DataFrame' object has no attribute 'as_matrix', By the way, thanks 💚 the tutorial is very clear and well explained . 👏Bravo

@vindyaabeygunawardena6222

2 жыл бұрын

change it into data.(['Review_class'].to_numpy().

how can we go back to the original sentence from X_test? I mean how can I see what sentences the algorithm doesn't classify correctly?

@lpereira95

3 жыл бұрын

Compare y_pred and y_test. The indices where they don't match are the mistaken samples. Then use that indices to access the sentences in X_test. I hope it helps.

We first split the data and do Preprocessing right why you perform on whole dataset

@NormalizedNerd

3 жыл бұрын

Yes, ideally we should first split then preprocess. Here, the text preprocessing will remain the same for both train and test set so I did them together. However, I also formed the BOW model on the whole data; which is not the correct way. We should build it only on the training set then apply it on the test set. I did it just to make things a little easier.

Is X_train the bag of words?

@NormalizedNerd

3 жыл бұрын

X_train is the feature matrix.

plz a make tutorial how nlp is work on Bangla text datasets.

@NormalizedNerd

3 жыл бұрын

Sure I'll...stay tuned!

why we use deimitor as '/t'

@NormalizedNerd

4 жыл бұрын

In the .txt file, the values are separated by tab('/t') just like the values are separated by a comma in a .csv file. In pandas we have the function read_csv(reads .csv files by default). We need to pass the parameter to read tab-separated files.

How can we access the text files used?

@NormalizedNerd

3 жыл бұрын

I've provide the link in the video description

One suggestion here. Please ZOOM-IN your screen while you are explaining the coding part. Press Ctrl and scroll up from your mouse, it will zoom in. It puts strain on our eyes and understanding the coding part becomes a punishment.

@NormalizedNerd

3 жыл бұрын

Point noted. Thanks for the feedback.

Is it 'Lov' or 'Love' the root?

@NormalizedNerd

3 жыл бұрын

Good question. The thing is...stemming should give us 'lov'. But, Porter stemmer gives us 'love'. I guess the reason lies in the details of Porter stemmer's implementation.

powerpoint could have been used for nice presentation

I would have loved it if it was good. 😀😅Is this positive?

@sujandutta8981

4 жыл бұрын

Such complex shit ;__;

ভাই আমিও বাঙালি.! ❤❤❤

@NormalizedNerd

4 жыл бұрын

বেশ ভালো লাগলো আপনার কমেন্টটি পেয়ে। চ্যানেলটিকে আপনার পরিচিতদের মধ্যে শেয়ার করার অনুরোধ রইল। ❤️