Natural Language Interaction

Recommended Prerequisites

Artificial Intelligence; Machine Learning; Programming, ideally in Python.

Teaching Methods

The course is structured around two types of activity:

- Classes

- Project

With two types of classes:

- Theoretical, for exposing theoretical materials and discussing examples;

- Laboratories, for carrying out practical exercises and supporting the development of the project.

The project will include a research component and the exploration of one or more approaches to solve a problem related to natural language interaction, also considering its evaluation.

Learning Outcomes

Acquisition of the following skills is expected:

- Methods for computational representation and manipulation of natural language;

- Application of various techniques for interaction in natural language, namely for searching documents, getting answers to questions, and in dialogue.

The acquisition of the following concepts is expected:

- Natural Language Processing

- Word and Sentence Embedding

- Information Retrieval

- Dialog Acts

- Dialog Flows

- Language Modeling

- Large Language Model

- Prompt Engineering

Work Placement(s)

No

Syllabus

1. Introduction to Natural Language Processing

- Levels

- Tasks

- Challenges

 

2. Vector Semantics

- Vector-space model

- Distributional Semantics

- Semantic Similarity

 

3. Automatic Question Answering

- Information Retrieval

- Semantic parsing

- Based on Knowledge

- Machine Reading Comprehension

 

4. Dialogue Systems

- Chat

- Task-oriented

- Corpora-based

- Dialogue Flows

- Dialogue State Tracking

- Other Natural Language Interfaces

 

5. Language Models

- Language Modeling

- Pre-training and Fine-tuning

- Use in zero mode and few-shot learning

- Prompt Engineering

- Applications

 

6. Evaluation

- Datasets

- Metrics

Assessment Methods

Assessment
Exam: 40.0%
Project: 60.0%

Bibliography

Jurafsky, D. and Martin, J. H. (2023). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition. Prentice Hall series in Artificial Intelligence. Prentice Hall, Pearson Education International, Englewood Cliffs, NJ, 3rd edition draft. 

Eisenstein, J. (2019). Natural Language Processing. MIT Press. 

Tunstall, L., von Werra, L., and Wolf, T. (2022). Natural language processing with transformers. O’Reilly Media, Inc.