Semantic and Natural Language Technologies

Recommended Prerequisites

Most desired non-essential skills:

- Artificial Intelligence Bases

- Computational Learning Bases

- Programming, ideally in Python or Java.

Teaching Methods

The course is structured around two main activities:

- Lectures

- Project

There will be two types of classes:

-Expositive, where the main concepts will be presented.

-Project support, in order to guide the students and assess progress.

The project includes a research component and a set of experiments that involve either: the creation or enrichment of a semantic database with knowledge extracted from text; exploiting one or more semantic databases in a natural language processing task.

Learning Outcomes

The following skills will be acquired:

-Using Semantic Web languages (RDF, RDFs, OWL) for data and knowledge representation.

-SPARQL for querying semantic databases.

-Reusing vocabularies and Linked Data.

-Foundations of Natural Language Processing.

-Creation and enrichment of semantic databases through natural language processing.

-Semantic databases in natural language processing and search.

-Semantic vector spaces and language models as an alternative to semantic databases.

Based on the previous, the following concepts will be acquired:

-Taxonomy, thesaurus, ontology, and other knowledge structures;

-Linked Data;

-Natural Language Processing;

-Information Extraction;

-Semantic Similarity;

-Distributional Semantics;

-Language Modeling;

-Information Retrieval;

-Semantic Search.

Work Placement(s)

No

Syllabus

1. Knowledge Representation:

            a. Languages RDF, RDF Schema, OWL

            b. Taxonomies, thesauri, ontologies and other knowledge representations

            c. Linked Data and Vocabulary Reutilization

            d. Querying with SPARQL

2. From text to structured knowledge:

            a. Natural Language Processing

            b. Information Extraction

            c. Lexical-Semantic Knowledge Bases

3. Distributional Semantics

            a. Vector Semantics

            b. Neural Language Models

4. Information Retrieval and Semantic Search

            a. In structured data

            b. In ustructured data (text).

Assessment Methods

Assessment
Exam: 35.0%
Project: 65.0%

Bibliography

-- Liyang Yu. A Developer's Guide to the Semantic Web, 3rd edition. Springer, 2015.

-- Dan Jurafsky & James H. Martin. Speech and Language Processing. Prentice Hall, 2009.

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

-- Diana Maynard, Kalina Bontcheva & Isabelle Augenstein (2016). Natural Language Processing for the Semantic Web. Morgan & Claypool Publishers.