Programme

Video recordings of the lectures are now available online.

Lectures

• (L1) Introduction to Linked Data
  Sören Auer, Axel-Cyrille Ngonga Ngomo, Jens Lehmann
  
• (L2) Foundations of Description Logics
  Sebastian Rudolph
  
• (L3) Using SPARQL with RDFS and OWL entailment
  Birte Glimm
  
• (L4) Scalable OWL2 Reasoning for Linked Data
  Aidan Hogan , Jeff Pan
  
• (L5) Rules and Logic Programming for the Web
  Adrian Paschke
  
• (L6) OWL and Rules
  Pascal Hitzler
  
• (L7) Models for the Web of Data
  Claudio Gutierrez
  
• (L8) Trust Management Methodologies for the Web
  Denis Trcek
  
• (L9) Scalable non-standard Reasoning on the Semantic Web
  Abraham Bernstein
  
• (L10) Database foundations for scalable RDF processing
  Katja Hose, Martin Theobald, Ralf Schenkel, Gerhard Weikum
  
• (L11) Combining Probabilistic and Logical Reasoning for Web Data Processing
  Heiner Stuckenschmidt, Mathias Niepert
  
• (L12) An Introduction to Constraint Programming and Combinatorial Optimisation
  Barry O'Sullivan, Helmut Simonis
  

Lecture Notes

As in previous years, we will publish lecture notes with Springer's Lecture Notes in Computer Science (LNCS) series and all students will receive a copy included in the registration fees.

Course Descriptions

Introduction to Linked Data (Sören Auer, Axel-Cyrille Ngonga Ngomo, Jens Lehmann)
[ PDF slides ]

With Linked Data, recently a very pragmatic approach towards achieving
the vision of the Semantic Web gained much traction. While many
standards, methods and technologies developed within the Semantic Web
activity are applicable for Linked Data, there are a also a number of
specific characteristics of Linked Data, which have to be considered.
In this article we introduce the main concepts of Linked Data.
We present an overview of the Linked Data life-cycle and discuss
individual approaches as well as the state-of-the-art with regard to
extraction, storage and querying, authoring, linking, enrichment,
quality analysis, evolution, as well as search and exploration of Linked
Data. We conclude the article with an discussion of issues, limitations
and further research and development challenges of Linked Data.

Outline of the Course:

• Introduction
• The Linked Data paradigm
• The Linked Data life-cycle
• Extraction
• Storage and Querying
• Authoring
• Linking
• Enrichment
• Quality Analysis
• Evolution
• Search and Exploration
• Conclusions, Outlook and Future Challenges

Foundations of Description Logics (Sebastian Rudolph)
[ PDF slides ]

Description Logics (DLs) constitute the basis of the Web Ontology Languages OWL 1 and OWL 2. DL research started as early as in the 1970s and by now has led to an impressive amount of theoretical results as well as several highly optimized tools for working with ontological knowledge.
The lecture will provide an overview of the state of the art in DLs and draw relations to other logical formalisms. We will introduce syntax and semantics and provide examples for modelling specific situations. We will review the most common approaches to reasoning and finally explain the tight correspondences between DLs and OWL.

Outline of the Course:

• What are DLs?
• Syntax and Semantics
• Modeling with DLs
• DL Reasoning approaches
• DLs and OWL

OWL and Rules (Pascal Hitzler)
[ Accompanying Tutorial Website (incl. slides) ]

The revision 2 of the Web Ontology Language OWL is much richer than its
predecessor OWL 1.0 with respect to modelling with rules. In particular,
a significant portion of OWL 2 DL is already expressible using rules.
OWL 2 can further be extended to cover DL-safe Datalog rules while
retaining decidability. This extension further more gives rise to a
tractable language, which covers (most of) all three tractable profiles
of OWL 2.

Using SPARQL with RDFS and OWL entailment (Birte Glimm)
[ PDF slides and handouts ]

SPARQL is a query language for RDF data and the main mechanism for
computing query results in SPARQL is subgraph matching: RDF triples in
both the queried RDF data and the query pattern are interpreted as
nodes and edges of directed graphs, and the resulting query graph is
matched to the data graph using variables as wild cards. Various W3C
standards, including RDFS and OWL, provide semantic interpretations for
RDF graphs that allow additional RDF statements to be inferred from
explicitly given assertions. In order to use SPARQL as a query
language in these cases as well, the currently devised SPARQL
entailment regimes define the evaluation of basic graph patterns using
semantic entailment relations instead of explicitly given graph
structures. In this lecture, we introduce the basics of SPARQL and
then introduce SPARQL's entailment regimes with a focus on basic graph
pattern evaluation in the SPARQL OWL entailment regime.

Outline of the Course:

• Introduction to SPARQL
• SPARQL Query Algebra
• Blank Nodes in SPARQL
• Subgraph matching via Simple Entailment
• Relationship between OWL DL and RDF
• The SPARQL OWL Entailment Regime
• Implementation Techniques

Scalable OWL2 Reasoning for Linked Data ( Aidan Hogan , Jeff Pan)
[ slides part 1 (.pptx | .pdf), slides part 2 ]

The goal of the Scalable OWL 2 Reasoning for Linked Data tutorial is twofold: first, to introduce scalable reasoning and querying techniques to SW researchers as powerful tools to make use of linked data and large-scale ontologies, and second, to present interesting research problems for SW that arise in dealing with TBox and ABox reasoning in OWL 2. The tutorial consists of three parts. It will begin with an introduction of linked data, as well as its relationship with ontologies, such as the ones in OWL 2. It will then start with the instance retrieval reasoning service, introducing how to provide scalable reasoning services for OWL 2 RL (a tractable fragment of OWL 2). The third part of the tutorial will present recent work on faithful approximate reasoning for OWL2-DL. The tutorial will include our implementation of the mentioned techniques as well as their evaluations.

Combining Probabilistic and Logical Reasoning for Web Data Processing (Heiner Stuckenschmidt,
Mathias Niepert )
[ PDF slides part 1, PDF slides part 2 ]

Data on the semantic web is often characterized by a combination of hard
logical definitions in terms of ontologies or linked data and unstructured
information in terms of textual descriptions or user generated contents.
While logical reasoning has proven useful as a tool for dealing with
ontologies and structured data, it often fails to capture the weakly
unstructured information. This kind of information is best handled by
probabilistic methods originating from information retrieval and machine
learning. In this course, we will show how Markov Logic, a combination of
first-order representations and probabilistic Markov networks can be used to
provide a unified framework for reasoning about combinations of hard and
soft knowledge on the semantic web, focusing on the problem of information
integration. Besides introducing Markov Logic and explaining its relation to
established semantic web languages, we will introduce typical semantic web
tasks, namely information extraction and information integration as examples
and present solutions to these problems based on Markov logic.

Outline of the Course:

• Introduction: Hard and Soft Knowledge on the Web
• Example Problems:
  • Information Extraction
  • Information Integration
• Markov Logic
  • Theory
  • Relation to Semantic Web Languages
• Using Markov Logic
  • Ontology Matching with Markov Logic
  • Object Reconciliation with Markov Logic
  • Ontology Learning with Markov Logic
• Conclusions

Rules and Logic Programming for the Web (Adrian Paschke)
[ PDF slides ]

This lecture will give an introduction to rule based knowledge
representation on Web. It will review the logical foundations of logic programming
and derivation rule languages and describes existing Web rule
standard languages, in particular the W3C recommendation Rule Interchange
Format (RIF).

Trust Management Methodologies for the Web (Denis Trcek)
[slides available to students upon their request via e-mail]

Trust management methodologies and technologies are becoming crucial for the further and wider acceptance of web based solutions. Trust related issues in e-environments were addressed for the first time some fifteen years ago, but the proposed approaches were actually tackling security and not trust. Later, more advanced methodologies emerged that were based on Bayesian statistics. These were followed by Dempster-Shafer theory of evidence and its derivatives. Further, game theory based methodologies were developed as well.
But trust in its core is reasoning, assessment process. Such mental assessment processes are governed by various factors and can be rarely assumed to be e.g. purely rational; trust management methodologies (and supporting technological solutions) should be aligned with this fact. This lecture will therefore provide an extensive overview of existing methodologies, followed by their evaluation, where focus will be on mentally ergonomic factors. Further, some latest experimental results will be given that identify and evaluate some of those most important factors. Finally, we will present one mentally ergonomic trust management methodology called Qualitative Assessment Dynamics (aka Qualitative Algebra, QA) that complements existing methodologies mentioned above, and that is aligned with the results of our experimental findings.

Outline of the Course:

• The importance of trust management in Web environments
• A short historical overview
• The main existing methodologies rooted in the computing and information science area
  • Naïve trust management
  • Subjective algebra
  • Game theory based approaches
• Analysis of existing approaches
  • Identification of open issues
  • Experimental verification of open issues
• Qualitative Assessments Dynamics (Qualitative Algebra)
• Conclusions and references

Models for the Web of Data (Claudio Gutierrez)

As people are publishing and querying structured data (as oposed to
documents) at Web scale, differents point of views and movements around
this phenomena have emerged. The most successful models today
are probably Linked Data, which generalizes the notion of link, and
Open data, which follows an infomation system approach. There have been,
nonetheless, other approaches to model the Web of Data. The goal of this
tutorial is to present these different models and briefly discuss from a
technical point of view the main characteristic of each of them.

Database foundations for scalable RDF processing (Katja Hose, Martin Theobald, Ralf Schenkel, Gerhard Weikum)
[ PDF slides part 1,PDF slides part 2, PDF slides part 3 ]

As more and more data is provided in RDF format, storing huge amounts of
RDF data and efficiently processing queries on such data is becoming
increasingly important.
The first part of the lecture will introduce state-of-the-art techniques
for scalably storing and querying RDF with relational systems, including
alternatives for storing RDF, efficient index structures, and query
optimization techniques.
As centralized RDF repositories have limitations in scalability and
failure tolerance, decentralized architectures, for
instance based on P2P systems, have been proposed. The second part of
the lecture will highlight system architectures, discuss index
structures, and introduce efficient techniques for distributed query
processing in general and for RDF data in particular.
Extracting knowledge from the Web is an excellent showcase -- and
potentially one of the biggest challenges -- for the scalable management
of uncertain data we have seen so far. The third part of the lecture is
intended to provide a close-up on current approaches and platforms to
make reasoning (e.g., in the form of probabilistic inference) with
uncertain RDF data scalable to billions of triples.

Outline of the Course:

• Relational RDF processing
  • Representing RDF in row stores and column stores
  • indexes and basic query processing techniques
  • staticics and query optimization
• Distributed RDF processing
  • System architectures
  • Index structures
  • Techniques for distributed query processing
• Scalable Uncertainty Management in RDF
  • Representation models for uncertain RDF
  • Distributed reasoning approaches
  • Approximations and guarantees

Scalable non-standard Reasoning on the Semantic Web (Abraham Bernstein)
[ PDF slides ]

As the web of linked data grows it becomes increasingly apparent that
the use of description logic might not always be the best way to get
good results. As a consequence other reasoning methods, such as
analogical or statistical reasoning, have become increasingly important.
This course will first cover the basics of of non-standard reasoning
methods for large-scale graphs. Next, it will then discuss what
frameworks are available to process large graphs -- contrasting their
strengths and weaknesses. To exemplify these, it will provide an
in-depth presentation of a parallelizing open-source graph-computation
framework. The lecture will close by showing how a series of
non-standard reasoning techniques can be elegantly implemented with the
framework in a scalable manner.

An Introduction to Constraint Programming and Combinatorial Optimisation (Barry O'Sullivan , Helmut Simonis)
[ PDF slides ]

Computers play an increasingly important role in helping individuals and
industries make decisions. For example they can help individuals make decisions
about which products to purchase or industries make decisions about how best to
manufacture these products. Constraint programming provides powerful support
for decision-making; it is able to search quickly through an enormous space of
choices, and infer the implications of those choices. This tutorial will teach
attendees how to develop models of combinatorial problems and solve them using
constraint programming, satisfiability and mixed integer programming techniques.
The tutorial will make use of Numberjack, an open-source Python-based optimisation
system developed at the Cork Constraint Computation Centre. The focus of the tutorial
will be on various network design problems and optimisation challenges in the web.

(Further details coming soon)