NLP Exercises: bag_of_n_grams and TF-IDF

[KirandeepMarala]

Hi sir,

Thanks for the suggestions to improve the sentences. Have done a thorough Grammarly check and corrected the sentence formations for bag_of_n_grams_exercise.

Have added the exercises files for TF-IDF, looking forward to feedback and any improvements to make it more better.

Thank you!

[dhavalsays]

"etc. for text representation and apply difference classification algorithms. " Text representation is better over pre-processing

[dhavalsays]

I suggest we import the libraries only when we need it. So you should start with "About Data" first and then whenever you need a library import that in that cell. In jupyter notebooks it is better to keep libraries near to cells where they are being used.

[dhavalsays]

Can you remove "perfectly" ? They are balanced but I would not call it perfectly balanced unless sample count is exactly same for all classes.

[dhavalsays]

Can you use lower case? i.e. preprocessed_comment ? It is recommended to use snake case and that too with everything as small case for column names, variables etc.

[dhavalsays]

Text should be text. Start with a small case. One observation I have Kiran is your English needs improvement. For this, you can use the Grammarly extension. In a language, the first letter of any sentence should be capitalized and then everything should be small with some exceptions such as the person's name, I, location, and so on. Read this for complete set of rules: https://writer.com/blog/capitalization-rules/

[dhavalsays]

Another example of improper capitalization. Please have all your text run through Grammarly check and make necessary changes. Overall once again good job on the actual technical content 👍