DBATU Final Year AI & ML syllabus

DBATU · Final Year B.Tech Artificial Intelligence & Machine Learning · Non-NEP

Natural Language Processing

BTAIC701 · Semester VII · 4 credits · Continuous assessment 20 + mid-semester 20 + end-semester 60

All notes for this subject
  1. Unit 1: Introduction to NLP

    7 hours

    The scope and applications of NLP; linguistic fundamentals, regular expressions, words and corpora; text normalisation: tokenisation, segmentation, stop words, stemming and lemmatisation; minimum edit distance.

    Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then

    • Levels and challenges of NLP
    • Regular expressions and finite automata
    • Corpora and annotation
    • Tokenisation and normalisation
    • Stemming and lemmatisation
    • Minimum edit distance
  2. Unit 2: Language Models and Vector Semantics

    8 hours

    N-gram models, their evaluation and smoothing; information retrieval; vector space models, TF-IDF and pointwise mutual information; word2vec; relevance ranking.

    Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then

    • N-gram language models
    • Perplexity and smoothing
    • TF-IDF and vector space models
    • Pointwise mutual information
    • Word representations
    • Information retrieval and extraction
  3. Unit 3: Sequence Labelling

    7 hours

    Text preprocessing and context-free grammars; part-of-speech tagging with HMMs and CRFs; named entity recognition and its evaluation; dependency and constituency parsing; maximum entropy Markov models.

    Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then

    • Context-free grammars
    • POS tagging
    • Hidden Markov models
    • Maximum entropy models and CRFs
    • Named entity recognition
    • Dependency parsing
    • Earley and CYK parsing
  4. Unit 4: Text Classification and Sentiment Analysis

    7 hours

    Classifiers for text classification and sentiment analysis; logistic regression for text, multinomial logistic regression, cross-entropy loss, gradient descent and regularisation; interpreting models.

    Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then

    • Naive Bayes text classification
    • Sentiment analysis
    • Logistic regression for text classification
  5. Unit 5: Deep Learning for NLP Applications

    7 hours

    Simple, stacked and bidirectional RNNs; LSTMs and GRUs; the encoder–decoder model; words, characters and byte pairs; transformers, pretrained and masked language models and fine-tuning; case studies from chatbots and summarisation to speech.

    Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then

    • Recurrent neural networks for NLP
    • LSTMs and GRUs for language
    • Encoder-decoder models
    • Subword tokenisation
    • Transformer architecture
    • Fine-tuning pretrained and masked language models
    • Question answering systems

Unit contents are summarised from the official DBATU syllabus . AiCanCode.org is not affiliated with Dr. Babasaheb Ambedkar Technological University (DBATU).