Text Classification | NLP Lecture 6 | End to End | Average Word2Vec

[@motivation_with_harsh]

you are the best teacher in this world sir best , teacher ever .........

[@piyushagarwal5314]

Hello Sir, please continue the playlist, Great/Best lectures! Anticipating second part of this text classification using Deep learning.

[@Ayaz-m5v]

sir mai fan hogya aapka mtlb ki mai wait krta hu jb bhi aapka koi naya lecture aata hai ,,i mean the way you teach is phenomenal....

[@k_abhishek2056]

Seriously i love ur explanation way..Dil se shukriya bhai aapko..lots of love and respect bhai apko❤🙌

[@pratheekv218]

Thank you sir for these course. Each lecture from you is very comprehensive and easy to understand. Note: In between POS tagging lecture is not been uploaded. It was Chapter 4 as you said in the introduction video.

[@asifsvirtualfootprint7370]

Great lecture. I hope we will get to see the second part of this text classification using Deep learning.

[@MrKB_SSJ2]

IT WAS AMAZING!!! I LEARNED SO MUCH FROM JUST THIS ONE VIDEO OF YOURS. THANK YOU!!!

[@kindaeasy9797]

at 37:38 i think the memory error is because you have a combined vocabulary of uni and bi gram which is obviously going to be huge , i think using just the bigram vocabulary by setting ngram_range=(2,2) will not cause this issue and your model will perform better

[@SiddheshDesai1993]

1:04 - Plan of Attack 3:34 - What is Text Classification 5:38 - Types of Text Classification 7:34 - Applications 12:36 - The Pipeline 17:56 - Different Approaches 21:32 - Heuristic Approach 23:16 - Using API 27:14 - Using BoW and n-grams 38:58 - Using Tf-Idf 40:44 - Using Word2Vec 48:33 - Practical Advise

[@akshaykumar192]

Ok thanks

[@vimmysaini]

Hello plz make a video on text summarization..... extraction method.

[@adityajaiswal9356]

❤️❤️You are a great teacher

[@Sandesh.Deshmukh]

Nice Video Nitish Sir... Actually we are waiting for complete NLP playlist. We all know because of Mentorship Program you are too busy, but we just want roadmap of Deep learning for NLP. So we can move forward with NLP. Thank you ❤❤

[@shivamsahu9998]

Awesome Tutorial

[@jade_warrior]

Sir do saal ho gaye, aur kitna intezar karwayenge Deep Learning implementation ke liye?

[@muhammadjaved9169]

you are best in india

[@Sandesh.Deshmukh]

Nice Video Nitish Sir... Actually we are waiting for complete NLP playlist. We all know because of Mentorship Program you are too busy, but we just want roadmap of Deep learning for NLP. So we can move forward with NLP. Thank you

[@ronylpatil]

Sir LSTM or RNN k videos bhi karo na upload, may be 50%+ logo ko deep learning phle se pta ho

[@SidIndian082]

Check in KZbin -- "KGP Taklie" , NLP playlist - LSTM for fake news vedio, concept & code is mentioned .... my rating is 5/5 for RNN you can to watch krish naik NLP playlist vedio (personally i hav'nt watched the vedio)..

[@narendraparmar1631]

Your videos are helping a lot. Thanks a lot sir ji😀

[@Gayu_2003]

Very nice explanation sir 👏🥰

[@charanpoojary4804]

Thank you sir

[@descendantsoftheheroes_660]

Great Job sir ..God bless you...webscrapping ,Tableau, Flask ,Python ye sb B tutorial kijiye ... HINGLISH TUTORIAL ME KHI NHI H..God bless sir to teaching well

[@campusx-official]

Sabke upar already hai. Check the channel

[@descendantsoftheheroes_660]

@@campusx-official web scraping ka link dena sir ...Mujhe nhi mila sorry sir

[@campusx-official]

@@descendantsoftheheroes_660 kzbin.info/www/bejne/bn-ylZqgeJZsbpo

[@descendantsoftheheroes_660]

@@campusx-official thanks sir love you 🤗🔥🔥

[@asifpervezpolok2243]

very informative .

[@GhostRider....]

Nice Teaching sir

[@riiyyyaaaa]

Sir, I have faced problem here while testing the model with first model then with second model . It s giving that the x_test features doesn't match the input features dimension . like 48k for i/p and 28k for test i/p. And the whole lecture is superb

[@saisrinivaspatnam15]

methods are actually different for test and train bow. cv.fit_transform for trainset and just cv.transform for testset.I too got trapped there.

[@harshitagupta5372]

Very informative video. I am new to this field. Sir, as a business/data analyst we validate business rule where in we have some business rules. For now we use sql case to check the completeteness and validity of the business rule. But in future we want to create python code to run the business rule automatically on data to check validity and completeness. Why do we need this, as the business rules for the attribute keep on changing so we just want python code read the rule from server and just tell us the completeness and validity of the attribute. Our business rule are like (must be populated if account in (reatail, saving) and has a lenght of 9 digit). I hope this is possible using nlp. If you have any suggestion please reply

[@rashadurrahman6182]

Sir, have you provided the "Deep Learning " version of text classification?

[@PANDURANG99]

How to do Multi Level Hierarchical Classification

[@NidhiSinha4U]

24:23, 36:40

[@joyeetamallik5063]

Hi Sir! Is Accuracy , Confusion matrix sufficient to judge the NLP classification models? or any other techniques also exists?

[@SidIndian082]

F1 score , precision call must

[@shivaaryaprakash]

Please upload deep learning approach wala video..

[@element6101]

LOVE YOU BRO...

[@AbhishekMishra-mq4jw]

Hi Nitish bhai ek video bnaado NER aur topic modelling pe bai bat complex topics hy sir k uprr se jata 😅😅 hy kuki ek interview merse ye poocha aur mei explain nhi kr paya dhngg se and got rejected its a request please🙏🙏

[@scchouhansanjay]

bhai, Onenote pe likhne ke liye kya use karte ho? purchase link de sakte ho pls?

[@zkhan2023]

Thanks Sir

[@prashantranjan7146]

where i can find aws google news vector link?

[@ohm1183]

Absolutely awesome video. How I can contact with you

[@thelife5628]

🔥🔥Thankyou sir

[@viratmani7011]

Sir I have a small doubt It would be great help if you help me in that. I had strucked figuring it out. In word2vec we have 2 vector representation for a word. When word is center word we have one representation and when it is context word we have another representation. Why do we have two vectors?

[@KumR]

Done

[@nirakaar3947]

Thank you for the amazing videos you are making and thanks for your efforts. I have 1 question, In your code X_train_bow = cv.fit_transform(X_train['review']).toarray() X_test_bow = cv.transform(X_test['review']).toarray() why cv.fit_transform and cv.transform differently used for train and test. I checked it changes the shape but what exactly is the reason. Kindly have a look at my question.

[@saisrinivaspatnam15]

once we fit to the trainset,we will transform the testset using the vocabulary of trainset so that there is dimensional consistency while we input to the GNB model for prediction

[@AyushSengar-z9z]

you havent used the story variable anywhere, why did you calculate it?

[@vanshikamittal4652]

i want to ask that what is the conclusion of result...it is showing accuracy of positive or negative reviews?

[@shuja5044]

If my sentiment column had multiple classes insted of binary, would LabelEncoder still work?

[@vijaybhatt8060]

thank you so much

[@tanmaythaker2905]

How can we handle imbalanced data in case of Text classification? Maybe machine translation? Or replacing words in the minority classes with synonyms. Any other techniques?

[@aarunibhugul]

I usually do 'representative sampling'. You have a look at it.

[@Sandesh.Deshmukh]

We are waiting for DL part Nitish Sir..Please make videos on how can we use DL for advanced DL.If this is not possible, please make roadmap so we can go with that. Thank you for efforts ❤❤

[@rajeevranjan5007]

Sir, the word "not" is also included in the english stopwords list. So do we need to first remove the word "not" from the stopwords list and then perform the stopwords removal step? OR is it fine to simply do the stopword removal as it for sentiment analysis? I have a doubt that the word "not" has the potential to disturb the accuracy of sentiment analysis task. Any Thoughts..?

[@campusx-official]

Depends on the application

[@piyushpathak7311]

Sir PLZZ 🙏 upload video of xgboost and DBSCAN algorithm and anomaly detection in ml playlist PLZZ 🙏 Sir

[@freshersadda8176]

Sir ml playlist me aur video add kr do

[@shashankpandey1966]

Sir total kitne lecture hai NLP playlist me ?

[@campusx-official]

4 aur

[@debojitmandal8670]

how to increase the accuracy please can u make a video on that

[@varunahlawat9013]

33:26 When I'm getting attribute error on running that cv.fit_transform(X_train['review']) I'm not even able to figure out, please help.

[@raghavjuneha148]

Simply do fit_transform X_train since review is not present in column , after use

[@ATHARVA89]

sir NLP series ka next video kyu nai aaraha?

[@campusx-official]

Kyuki ml roadmap wala video banane me time nai mila. Iss week aa jaayega

[@Ankit-hs9nb]

21:17 is the part 2 deep learning video out?

[@karthiavenger4577]

sir please complete deep learning playlist please

[@chauhanabhishek9593]

❤️

[@nanditagautam6310]

sir plz make a video on BERT . please

[@abhishekgandhi174]

How should we reach you ?? We wanted to have mentorship

[@rahulkumar-op7ul]

Thank you sir

[@ShivaniSharma-tk4bl]

Plz share the link to the codes

[@poojawasnik7617]

Sir plz provide part 2

[@avnigoyal7736]

X = [] for doc in tqdm(df['review'].values): X.append(document_vector(doc)) ValueError: need at least one array to concatenate while running this code for my whole dataset this error is coming can someone help

[@longtreegreen]

Sir why do you make noises like chap-chup with tongue and nose. Its irritating.

[@dheerajkumar824]

4 min to bakchodi me waste kr di. Aise me views km ho jate h

[@ansylpinto2301]

Be polite sir. He is teaching quality content for free. He does not owe us anything,