Module 10: Classical NLP Approaches

• Cleaning & preprocessing natural language
• Tokenization strategies
• Building text features: BoW, TF-IDF
• Performing Sentiment Analysis & Spam Detection
• Named Entity Recognition

Module 11: Embeddings & Representation Learning

• Word2Vec (CBOW & Skip-Gram)
• GloVe, FastText
• Contextual embeddings (ELMo, BERT)
• Sentence-level embeddings
• Subword tokenization frameworks

Module 12: RNN, LSTM & GRU Ecosystem

• Sequential modelling concepts
• Handling the vanishing gradient problem
• Building predictive text models using LSTM
• When to use GRU vs LSTM
  
  Module 13: Seq2Seq & Attention Frameworks

• Encoder–Decoder workflows
• Attention mechanism explained simply
• Applications like translation & summarization
• Beam search for sequence generation

Module 14: Transformers — Complete Breakdown

• Self-attention explained visually
• Multi-head attention & positional encoding
• Modern transformer architectures:
• BERT
• GPT models
• RoBERTa
• T5 & BART
• DistilBERT & ALBERT