Natural Language Processing

Recommended Prerequisites

Artificial Intelligence Topics, Linear Algebra, Machine Learning

Teaching Methods

This Curricular Unit will have two types of classes:
- Lectures, where theoretical concepts will be presented and some practical examples discussed;
- Practical classes, for the familiarisation with computational tools, their application to small examples related with the theoretical concepts, which may also be used for project development support.
The project will consist of the development of an application / model for a NLP task, as well as its validation.

Learning Outcomes

Human language is the most natural way of communication. A great amount of data produced everyday (e.g. on the Web) are written in this form.

This unit will contribute with the acquisition of skills for (pre-)processing and manipulating this kind of data, given its specificities. This will be focused on knowledge and application of computational techniques for exploring and formalizing natural language, including:

-              Lexical, syntactic and semantic analysis;

-              Symbolic approaches as well as statistical processing;

-              Exploitation of available linguistic resources and tools;

-              Higher-level tasks dealing with natural language;

-              Extraction of features that may be explored by other intelligent systems, in prediction or decision-support tasks;

-              More theoretical visions and some linguistic knowledge.

Work Placement(s)

No

Syllabus

1. Introduction to Natural Language Processing
1.1 Formal languages and natural languages
1.2 Ambiguity, linguistic variability and other phenomena
1.3 Knowledge Levels
1.4 NLP Pipeline
2. Words
2.1 Resources: Lexicons and Corpora
2.2 Morphology Analysis
2.3 N-grams and Language Models
2.4 Sequence Processing for Part-of-Speech Tagging and Named Entity Recognition
3. Syntax
3.1 Formal Grammars
3.2 Parsing
3.3 Dependency Parsing
4. Semantics
4.1 Representation of Sentence Meaning
4.2 Lexical Semantics
4.3 Vector Semantics
4.4 Semantic Parsing
4.5 Correference Resolution
4.6 Word sense disambiguation and Entity Linking
5. Applications
5.1 Text Classification
5.2 Information Retrieval
5.3 Information Extraction
5.4 Automatic Question Answering
5.5 Dialog Systems
5.6 Other applications

Head Lecturer(s)

Catarina Helena Branco Simões da Silva

Assessment Methods

Assessment
Project: 50.0%
Exam: 50.0%

Bibliography

Dan Jurafsky and James H. Martin. Speech and Language Processing, Prentice Hall, 2019 (3rd edition draft).

 

Jacob Eisenstein. Natural Language Processing. MIT Press (draft edition, November, 2018).