• Credit Structure: 0.5-0-0-2-1

This course provides an in-depth understanding of the syntactic and semantic foundations of Natural Language Processing. It introduces the principles of grammatical structure, parsing algorithms, and meaning representation in language. Students will learn how syntactic and semantic analyses enable core NLP applications such as question answering, information extraction, and text understanding. Emphasis is placed on both classical and neural approaches to parsing and semantic role labeling.

Unit 1: Connecting Speech & Language 1 (1 hour)

• • Application overview connecting speech & language

Unit 2: Introduction to Speech Processing  (3 hours)

• • Speech Processing in Conversational AI, Speech Production, Spoken Language Processing

Unit 3: Speech Representation (5 hours)

• • Speech Signal, Sampling, Quantization, Frequency, Short-term Processing, Mel Spectrogram 

  • Speech Information Sources: Prosody, tone, emotion, accent, speaker, noise, sound pressure level.

Unit 4: Speech Representations (4 hours)

• • CNN, Transformer encoder, Wav2Vec2.0 encoder, Whisper encoder.

Unit 5: Lexical Processing in NLP (3 hours)

• • Language Modeling: N-grams, Word2Vec, GloVe.

Unit 6: Syntactic Processing in NLP  (6 hours)

• • Sequence Labeling for Parts-of-Speech (POS) Tagging and Named Entity Recognition (NER). Context-Free Grammars and Constituency Parsing.

  • Dependency Parsing.

Unit 7: Semantic Processing in NLP  (3 hours)

• • Word Sense Disambiguation (WSD), Semantic Role Labeling (SRL)

Unit 8: Connecting Speech & Language 2 (3 hours)

• • Application building connecting speech & language

Course Introduction 

• Lecture 1: Introduction to Syntax and Constituency Grammar

• Lecture 2: Parse Trees and Derivations 

• Lecture 3: Parsing Algorithms: Top-Down, Bottom-Up, and CYK

• Lecture 4: Probabilistic Context-Free Grammars (PCFGs)

• Lecture 5: Introduction to Dependency Grammar

• Lecture 6: Dependency Relations and Tree Constraints

• Lecture 7: Transition-Based Dependency Parsing

• Lecture 8: Graph-Based Dependency Parsing and Neural Parsers

• Lecture 9: Predicate-Argument Structures and Thematic Roles

• Lecture 10: Approaches to SRL: Rule-Based vs. Statistical

• Lecture 11: Applications of SRL in Question Answering and Information Extraction

• Lecture 12:  Introduction to Word Sense Disambiguation (WSD)

• Lecture 13: Lesk Algorithm and Semantic Similarity Approaches

• Lecture 14: Evaluation Frameworks and Case Studies

• Lecture 1: Application overview connecting speech & language

• Unit 2: Introduction to Speech Processing  (3 hours)

• Speech Processing in Conversational AI, Speech Production, Spoken Language Processing

• Unit 3: Speech Representation (5 hours)

• Speech Signal, Sampling, Quantization, Frequency, Short-term Processing, Mel Spectrogram 

• Speech Information Sources: Prosody, tone, emotion, accent, speaker, noise, sound pressure level.

• Unit 4: Speech Representations (4 hours)

• CNN, Transformer encoder, Wav2Vec2.0 encoder, Whisper encoder.

• Unit 5: Lexical Processing in NLP (3 hours)

• Language Modeling: N-grams, Word2Vec, GloVe.

• Unit 6: Syntactic Processing in NLP  (6 hours)

• Sequence Labeling for Parts-of-Speech (POS) Tagging and Named Entity Recognition (NER). Context-Free Grammars and Constituency Parsing.

• Dependency Parsing.

• Unit 7: Semantic Processing in NLP  (3 hours)

• Word Sense Disambiguation (WSD), Semantic Role Labeling (SRL)

• Unit 8: Connecting Speech & Language 2 (3 hours)

• Application building connecting speech & language

"Speech and Language Processing" by Daniel Jurafsky and James H. Martin, Prentice Hall, 2024.

"Natural Language Understanding" by James Allen, Benjamin/Cummings Publishing Company, 1987.

"Foundations of Statistical Natural Language Processing" by Christopher D. Manning and Hinrich Schütze, MIT Press, 1999.

"A Primer on Neural Network Models for Natural Language Processing" by Yoav Goldberg, Online.

"Natural Language Processing with Python" by Steven Bird, Ewan Klein, Edward Loper, O'Reilly Media, Inc., 2009.

• 2 Theoretical Assignments (14%)

• 12 Quizzes (36%)

• 1 End Term (30%)

• Classroom Notes (10%)

• Attendance (10%)