Natural language processing (NLP) Basics

So why NLP is imp because it can be used with machine learning and it can be used with deep learning

So what is exactly NLP is when our data is consist word, text, sentenses and paragraph then our ML model or deep learning will just will not able to understand those text so we need to convert those text into vectors there are variour process or ways to convert words or text into vectors so initially there are certain steps with respect to NLP

So there are various libraries for NLP which are as follows

1. spaCy
2. Natural Language Analyses with NLTK from Stanford university

In case of Deep Learning Section following libraries

1. PyTorch
2. Keras
3. TensorFlow

Steps for NLP :-

1. Text Processing Level -1

   - Tokenization

   - Lemmanization

   - Stop Words

2. Text Processing Level -2

   - Bag of Words

   - TFIDF

   - Unigrams

   - Bigrams

   - n-grams

   so in this step we converts words of sentenses into vectors with the help of above methods, so when we convert words into vectors then n then only our ML algorithm will be able to understand the context of that specific sentence

3. Text Processing

   - Genism

   - Word2vec

   - AvgWord2vec

   so above are some of the advance techniques for converting words into vectors in case of bag of words or TFIDF there are some flaws but with \ the help of above our performance with the respect to Text Processing will automatically inhance

4. Machine Learning Useacases

   after completion of above 3 steps we can do some of the ML usedcases

   - Sentiment Classifiers

   - Spamham Classifiers

   - Document Classifiers

5. Get the understanding of Artificial Nueral Network

   - Prerequisite

   - Basic Undestanding Of Deep Learning Techniques

   - Gradient Descents

6. Understanding Recurrent Neural Network, LSTM, GRU

7. Text Processing Level - 3

   - Word Embeddings

   - Word2vec

8. Bidirectional LSTM RNN, Encoders, And Decoders, Attention Models

9. Transformers

10. BERT