KeynotesSTAF 2024

We are thrilled to announce the invited speakers from across the STAF 24 conferences:

WADT’24 #1: Bernhard Möller, Augsburg University (Germany)

Some Uses of Modal Semirings

Monday 8 July, Waaier 3, 09:00

Abstract: We survey one branch of algebraic logic, namely modal semirings. They provide compact algebraic definitions of actions, with choice + and sequential composition . , together with multi-modal operators box and diamond that allow reasoning about successors and predecessors of states/worlds. Particular instances are homogeneous binary relations or sets of finite and infinite non-empty traces under fusing concatenation.

As main examples of applications we present obstacle analysis for geographic wayfinders, Hoare Logic, O’Hearn’s Incorrectness Logic, General Correctness Logic, as well as the temporal logic CTL^* and its sublogics CTL and LTL. We also give glimpses at Epistemic Logics of belief and knowledge, pointer structures and Separation Logic, and preference database queries.

WADT’24 #2: Jan A. Bergstra, University of Amsterdam (The Netherlands)

Common Meadows with Infinite Values

Monday 8 July, Waaier 3, 15:30

Abstract: Common meadows are enlargements of a field with an error value and a division function which is made total by setting 1/0 equal to said error value. We discuss properties and axiomatisations of common meadows, and of signed common meadows.

Moreover, we discuss signed infinite values which deviate from the infinite value of Anderson’s transrationals so that fracterm flattening is preserved. Using a modified quotient construction a transformation to a different enlargement of signed common meadows with infinite values.

ECMFA’24: Agnes Koschmider, Bayreuth University (Germany)

From Data Chaos to Decision Making

Tuesday 9 July, Waaier 2, 09:00

Abstract: This talk addresses how to efficiently process unstructured data for process mining. The volume of data is continuously increasing and the ability and demand to efficiently analyze the data has become even more crucial. Although several suitable techniques and tools already exist to efficiently process and analyze unstructured data, the challenge still exists how to intervene process orientation into unstructured data analysis. This combination promises uncovering new insights in terms of causal effects or bottlenecks in data that could not be directly found with alternative technique. Finally, involving users in such an analytics pipeline gives confidence in decision-making. This talk summarizes challenges, presents use cases, and gives an outlook on prospective research projects for process mining on unstructured data.

ICGT’24 #1: Mariëlle Stoelinga, University of Twente (The Netherlands)

Graphs, Logics and Transformations for Effective Risk Analysis

Wednesday 10 July, Waaier 2, 09:15

Abstract: Risk management is a fundamental process to ensure the reliable operation of systems, services, processes, and missions in our society. Examples range from self-driving cars, power grids, credit card payments, and military missions. Proper risk management techniques enable organizations to achieve their goals in an effective way and take effective mitigating measures.

Risk models support the risk management process in the identification, priorisation, and quantification of risks via effective preventive and corrective actions. Numerous industrial risk models exist. In this talk, I will focus on fault trees and attack trees, which are both top-down models that break high-level system risks into their causes, until the root causes are found. While fault trees focus on safety risks, i.e., unintended failures, attack trees take into account security risks, i.e., disruptions due to malicious attacks.

In this talk, I will take a graph-theoretic perspective on fault trees, attack trees, and their combination.

• First, I will present a formal semantics, which is surprisingly intricate given the fact that there are only a handful of logical gates to propagate failures and attacks.
• Next, I will propose several algorithms to analyse quantitative attack trees, based on BDDs and stochastic model checking, highlighting the role of graph transformations to make this process more efficient.
• Finally, I will present risk query logics, which allows engineers to query large attack and fault tree models.

Together these ingredients allow organisations to make better decisions on mitigating measures, making decisions more systematic, transparent and evidence-based–the increased constraints imposed by international standards, together with the ever-growing penetration of AI components in high-tech systems make rigorous and powerful risk management more important than ever.

MeSS’24: Giancarlo Guizzardi, University of Twente (The Netherlands)

It’s Patterns all the Way Down: Patterns, Anti-Patterns and Pattern Languages for Next-Generation Semantic Modelling

Wednesday 10 July, Waaier 3, 11:00

Abstract: Perhaps the most fundamental notion for successful large-scale collaboratively-built systems is Semantic Interoperability. This talk defends the view that to meet the requirements for full semantic interoperability, we need to create an engineering discipline that is based on the followings tenets:

• (i) real-world semantics should take precedence over pure formal semantics;
• (ii) real-world semantics is very much about “Ontology” in the true original sense of the term;
• (iii) before considering automated reasoning and computational non-functional requirements (e.g., computational tractability), semantic representation systems must strive for precision, expressivity, truthfulness to the underlying conceptualisation, and pragmatic efficiency;
• (iv) we need proper foundational theories and engineering (methodological, computational) tools based on these.

The talk presents the results of a two-decades research program aimed at addressing these issues. The program is based on two fundamental pillars: the development and use of foundational axiomatic theories (foundational ontologies) and principles in the design of semantic models (domain ontologies, knowledge graphs), and the use of patterns as central complexity management tools in this process. After introducing an Ontology Pattern Language (OPL) for Semantic Modelling that is truly ontology-based, I will discuss Ontological Anti-Patterns (OAPs) as fundamental tools for model validation. Moreover, I will elaborate on the role of having execution semantics of models for model understanding, and validation, and how this approach associated with a particular machine learning technique can support sustainable Anti-Pattern detection and rectification as well as model evolution.

AgileMDE’24: Iván Alfonso, Luxembourg Institute of Science and Technology (Luxembourg)

Low-Code and Low-Modeling Strategies for Agile MDE Processes

Wednesday 10 July, Waaier 3, 13:30

Abstract: The rise of agile methodologies has reshaped the landscape of software development and operation since the inception of the Agile Manifesto in 2001. Today, novel approaches are emerging to synergise agile methodologies with Model-Driven Engineering (Agile MDE), presenting additional advantages in terms of development efficiency, software quality, and collaborative workflows.

However, the adoption of MDE can pose challenges due to its demand for specialised techniques, knowledge, and tools, creating barriers in certain fields and industries. In this talk, we delve into the synergy of Low-Code and Low-Modeling to support and streamline stages of the development cycle of applications following the MDE approach.

Low-Modeling is focused on supporting the design of model, enabling diverse user participation in the application development process. Simultaneously, Low-Code, sharing a similar objective but focused on code, minimises the need for manual coding by automating the code generation process. Together, these approaches not only streamline development processes but also address the challenge of adopting Agile MDE by reducing complexity.

ICGT’24 #2: Tiago Prince Sales, University of Twente (The Netherlands)

Ontological foundations for building knowledge graphs

Thursday 11 July, Waaier 2, 09:00

Abstract: Knowledge graphs (KG) are gaining popularity in a wide range of domains and applications, both in industry and academia. A popular way to build them is by using the W3C Semantic Web standards, such as the Resource Description Framework (RDF) and the Web Ontology Language (OWL). A crucial step in the engineering of a KG is the design of the ontology that will be used to annotate its nodes and edges to provide real-world semantics. In this talk, I will discuss and illustrate how designing the right ontology for your KG presents both ontological and engineering challenges. The former includes those related to creating a model that accurately reflects how a community understands a domain of interest, while the latter consists of coming up with solutions to address the expressivity limitations of the languages we use. As an answer to these challenges, we have created gUFO, an OWL implementation of the Unified Foundational Ontology. In the talk, I will discuss how gUFO brings together a number of micro-theories from philosophy, logic, and cognitive science to offer knowledge engineers a foundation upon which they can build domain ontologies in a faster and more principled way.

LLM4MDE’24: Juri Di Rocco, University of L’Aquila (Italy)

Harnessing Large Language Models for Model-Driven Engineering: Current Applications and Future Directions

Thursday 11 July, Waaier 3, 11:15

Abstract: Model-Driven Engineering (MDE) has advanced significantly with the integration of Machine Learning (ML) and Deep Learning (DL) techniques. To understand the transformative potential of Large Language Models (LLMs) in MDE, we first explore the spectrum of artificial intelligence: from Artificial Narrow Intelligence (ANI), which specializes in specific MDE tasks, to the more versatile Artificial General Intelligence (AGI).

We will look at an overview of current LLM applications in MDE, emphasizing their role in automating tasks such as model repository classification and developing sophisticated recommender systems.

Furthermore, we provide a comprehensive outline of the technical considerations for integrating LLMs into MDE workflows. This practical guide is designed to equip researchers and practitioners with the necessary knowledge to seamlessly implement LLMs into their MDE processes.

Looking ahead, we will propose a targeted research agenda to explore the future interplay of LLMs and MDE, identifying key challenges and opportunities. This roadmap envisions the use of LLM techniques to enhance the management, exploration, and evolution of modeling ecosystems.

Accepted Papers