-
Presentation
Presentation
The Foundations of Natural Language Processing (NLP) module offers an in-depth exploration into the world of computational linguistics, aiming to provide students with a robust understanding of the tools, techniques, and models integral to analysing and processing human language data. Spanning 15 weeks, this course progresses from foundational topics, such as text normalization and regular expressions, to advanced themes in NLP, like vector semantics and the revolutionary transformer architectures, including chatGPT prompt engineering.
-
Class from course
Class from course
-
Degree | Semesters | ECTS
Degree | Semesters | ECTS
Master Degree | Semestral | 7
-
Year | Nature | Language
Year | Nature | Language
2 | Mandatory | Português
-
Code
Code
ULHT6347-26022
-
Prerequisites and corequisites
Prerequisites and corequisites
Not applicable
-
Professional Internship
Professional Internship
Não
-
Syllabus
Syllabus
Introduction and basic concepts Regular expressions and text normalization Bag-of-words Topic modelling Part I: NMF and LDA Bag-of-words Topic modelling Part II: Stochastic Block Modelling Naive Bayes and Sentiment Classification Vector semantics and embeddings Word2Vec Transformers and pre-trained language models Fine-tuning and masked language models LLM Prompting
-
Objectives
Objectives
Develop a foundational understanding of the basic concepts and challenges in NLP. Acquire proficiency in utilizing regular expressions for text processing and normalization. Gain insight into various topic modelling techniques, including NMF, LDA, and others. Understand the principles behind Naive Bayes and its application in sentiment classification. Dive deep into vector semantics, understanding word embeddings and their significance. Grasp the mechanics and implementation of Word2Vec. Learn about the transformative power of transformer architectures and pre-trained models in NLP. Become adept in fine-tuning, masked language models, and the art of prompting and instruct tuning for specific tasks.
-
Teaching methodologies and assessment
Teaching methodologies and assessment
The Foundations of NLP module employs a blend of theoretical instruction and hands-on practice to foster comprehensive learning. Lectures will be conducted to elucidate concepts, models, and techniques, supplemented with real-world examples and case studies. To enhance practical understanding, each lecture will be accompanied by tutorials that offer students opportunities to apply what they've learned, work on projects, and solve problems in a guided environment. These tutorials, integral to the learning process, ensure that students not only grasp the theoretical underpinnings but also gain practical experience, equipping them for both academic research and industry applications.
-
References
References
Jurafsky, D., & Martin, J. H. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition. 3rd Edition
-
Office Hours
Office Hours
-
Mobility
Mobility
No
