NLP

Natural-Language-Processing

Week 1

Text Preprocessing Techniques and Basics for NLP

• Regularization, Text Normalization, stemming, lemmatiziation, tokenization and other basics of NLP.

  • Watch the videos from video 1 to 13 by TO courses on YouTube.
  • Read the 2nd chapter on Regularization, Text Normalization and Edit Distance by Stanford.edu.
  • Read the blog post on Ultimate Guide to Understand and Implement Natural Language Processing with codes in Python by Analytics Vidhya.
  • Read my blog post on Natural Language Processing:-A Beginner's Introduction
    • See my Jupyter NoteBook on the Beginner's introduction for the code.

• Assignment

  • Use NLTK to perform stemming, lemmatiziation, tokenization, stopword removal on a dataset of your choice.

Week 2

Word Embeddings in Natural Language Processing

• Embedding Layers, Word2vec, GloVe

  • Read the Blog Post on A Gentle Introduction to Statistical Language Modeling and Neural Language Models by Machine Learning Mastery.

  • Read the Material on Word2Vec Tutorial - The Skip-Gram Model by Stanford.edu CS224N

    • Read the paper titled Efficient Estimation of Word Representations in Vector Space

  • Read the Material on GloVe: Global Vectors for Word Representation by Stanford.edu CS224N

    • See my Jupyter Notebook on Word Embeddings for the code

• Assignment

  • 3 Assignments Visualize and Implement Word2Vec, Create dependency parser all in PyTorch (they are assigments from the stanford course)

Week 3

Lexicons and Language Models (Pre-Deep Learning)

• Lexicons and Pre-deep learning Statistical Language model pre-deep learning ( Hidden Markov Models, Topic Modeling with Latent Dirichlet Allocation)

  • Read the materials by CSEP 517: Natural Language Processing:-University of Washington, Spring 2017. Read the Lectures from Text Classifiers to Machine Translation(Lectures 2 to 6).
  • Read Your Guide to Latent Dirichlet Allocation by Lettier.
  • Watch the video on Topic Modeling with LDA on Youtube

Week 4

Deep Sequence Models

• Sequence to Sequence Models,(translation, summarization, question answering), Attention based models and Deep Semantic Similarity

  • Watch the course Natural Language Processing by Coursera, Week 4.
  • Read the blog post on DSSM (Deep Semantic Similarity Model) - Building in TensorFlow by Kishore P.V.
  • Read Chapter 10- Sequence Modeling, Recurrent and Recursive Nets of the Deep-Learning book by Ian Goodfellow

• Assignment

  • 3 Assignments, create a translator and a summarizer. All sequence to sequence models. In PyTorch.

Week 5

Dialog Systems and Transfer Learning

• Speech Recognition, Dialog Managers, Natural Language Understanding and Transfer Learning

  • Watch the course Natural Language Processing by Coursera, Week 5.
  • Read the Material on Dialog Systems and ChatBots by stanford.edu
  • Read the blog post on NLP-Imagenet by Sebastian Ruder
  • Read the blog post on Generalized Language Models by Lilian Weng.
  • Watch Siraj Raval's video How to Build a Biomedical Startup on Youtube
  • Read Transfer learning with BERT/GPT-2/ELMO by Jay Alammar

• Assignment

  • Try and Replicate my Project TextBrain:-Building an AI startup using Natural Language Processing
  • Play with Hugging Face:Pytorch Transformers pick 2 models, use it for one of 9 Down-Stream tasks, compare their results.

For more resources visit here