Now showing 1 - 10 of 401
  • Publication
    Metadata only
    Ordinal motifs in lattices
    (2024-02-01)
    Hirth, Johannes 
    ;
    Horn, Viktoria 
    ;
    Stumme, Gerd 
    ;
    Lattices are a commonly used structure for the representation and analysis of relational and ontological knowledge. In particular, the analysis of these requires a decomposition of a large and high-dimensional lattice into a set of understandably large parts. With the present work we propose /ordinal motifs/ as analytical units of meaning. We study these ordinal substructures (or standard scales) through order-embeddings and (full) scale-measures of formal contexts from the field of formal concept analysis. We show that the underlying decision problems are NP-complete and provide results on how one can incrementally identify ordinal motifs to save computational effort. Accompanying our theoretical results, we demonstrate how ordinal motifs can be leveraged to achieve textual explanations based on principles from human computer interaction.
      4
  • Publication
    Metadata only
    On the Challenges of Transforming UVL to IVML
    (2024)
    Agarwal, Prankur 
    ;
    Feichtinger, Kevin 
    ;
    ; ;
    Rabiser, Rick 
    Software product line techniques encourage the reuse and adaptation of software components for creating customized products or software systems. These different product variants have commonalities and differences, which are managed by variability modeling. Over the past three decades, both academia and industry have developed numerous variability modeling methods, each with its own advantages and disadvantages. Many of these methods have demonstrated their utility within specific domains or applications. However, comprehending the capabilities and differences among these approaches to pinpoint the most suitable one for a particular use case remains challenging. Thus, new modeling techniques and tailored tools for handling variability are frequently created. Transitioning between variability models through transformations from different approaches can help in understanding the benefits and drawbacks of different modeling approaches. However, implementing such transformations presents challenges, such as semantic preservation and avoiding information loss. TRAVART is a tool that helps with transitioning between different approaches by enabling the transformation of variability models into other variability models of different types. This paper discusses the challenges for such transformations between UVL and IVML. It also presents a one-way transformation from the UVL to IVML with as little information loss as possible.
  • Publication
    Metadata only
    Towards Ordinal Data Science
    (2023-12-19)
    Stumme, Gerd 
    ;
    Dürrschnabel, Dominik 
    ;
    Order is one of the main instruments to measure the relationship between objects in (empirical) data. However, compared to methods that use numerical properties of objects, the amount of ordinal methods developed is rather small. One reason for this is the limited availability of computational resources in the last century that would have been required for ordinal computations. Another reason - particularly important for this line of research - is that order-based methods are often seen as too mathematically rigorous for applying them to real-world data. In this paper, we will therefore discuss different means for measuring and ‘calculating’ with ordinal structures - a specific class of directed graphs - and show how to infer knowledge from them. Our aim is to establish Ordinal Data Science as a fundamentally new research agenda. Besides cross-fertilization with other cornerstone machine learning and knowledge representation methods, a broad range of disciplines will benefit from this endeavor, including, psychology, sociology, economics, web science, knowledge engineering, scientometrics.
      4
  • Publication
    Metadata only
    Divergent selection in a Mediterranean pine on local spatial scales
    (2023-11-27)
    Budde, Katharina B. 
    ;
    Rellstab, Christian 
    ;
    Heuertz, Myriam 
    ;
    Gugerli, Felix 
    ;
    ;
    Pausas, Juli G. 
    ;
    González‐Martínez, Santiago C. 
      4
  • Publication
    Metadata only
    Research topic flows in co-authorship networks
    (2023-09-01)
    Schäfermeier, Bastian 
    ;
    Hirth, Johannes 
    ;
    In scientometrics, scientific collaboration is often analyzed by means of co-authorships. An aspect which is often overlooked and more difficult to quantify is the flow of expertise between authors from different research topics, which is an important part of scientific progress. With the Topic Flow Network (TFN) we propose a graph structure for the analysis of research topic flows between scientific authors and their respective research fields. Based on a multi-graph and a topic model, our proposed network structure accounts for intratopic as well as intertopic flows. Our method requires for the construction of a TFN solely a corpus of publications (i.e., author and abstract information). From this, research topics are discovered automatically through non-negative matrix factorization. The thereof derived TFN allows for the application of social network analysis techniques, such as common metrics and community detection. Most importantly, it allows for the analysis of intertopic flows on a large, macroscopic scale, i.e., between research topic, as well as on a microscopic scale, i.e., between certain sets of authors. We demonstrate the utility of TFNs by applying our method to two comprehensive corpora of altogether 20 Mio. publications spanning more than 60 years of research in the fields computer science and mathematics. Our results give evidence that Topic Flow Networks are suitable, e.g., for the analysis of topical communities, the discovery of important authors in different fields, and, most notably, the analysis of intertopic flows, i.e., the transfer of topical expertise. Besides that, our method opens new directions for future research, such as the investigation of influence relationships between research fields.
      4
  • Publication
    Open Access
    Investigation of explainable AI (XAI) in commercial IIoT platforms
    (2023-07-06)
    Swora, Marlo 
    In the last years, artificial intelligence and machine learning algorithms are rising in importance and complexity. To increase the trust in these algorithms, they have to be as transparent as possible. Especially decisions of deep neural networks and similar complex black-box models are hard to explain and offer little insight. Explainable artificial intelligence (XAI) is a field of AI which tries to make complex AI models and their predictions interpretable. Currently, there are legislative changes at national and European level that require XAI as a prerequisite for artificial intelligence algorithms. This work provides an overview of some of the most relevant techniques of XAI and their use-cases, which can help to improve the transparency of complex AI models or boost the effectiveness of simpler interpretable AI models like Decision Trees. The report also provides an overview of the most established Industrial Internet of Things (IIoT) AI platforms regarding XAI and highlights their strengths and weaknesses in this area. This should make it easier to identify relevant XAI techniques while pointing to appropriate AI platforms.
      28  65
  • Publication
    Metadata only
    Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications
    The Asset Administration Shell (AAS) is an upcoming information model standard, which aims at interoperable modeling of “assets”, i.e., products, machines, services or digital twins in IIoT/Industry 4.0. Currently, a number of IIoT-platforms use proprietary information models similar to AAS, but not a common standard, which affects interoperability.A key question for a broad uptake is if AAS can be applied in a performant and scalable manner. In this paper, we examine this question for the open source Eclipse BaSyx middleware. To explore capabilities and possible performance limitations, we present four experiments measuring the performance of experimental AAS in BaSyx and, within the context set by our experiments, i.e., 10-1000 AAS instances, can conclude good scalability.
      11
  • Publication
    Metadata only
    Software in Cyberphysischen Produktionssystemen
    (2023)
    Feichtinger, Kevin 
    ;
    Meixner, Kristof 
    ;
    Rinker, Felix 
    ;
    Koren, Istvan 
    ;
    ;
    Heinemann, Tonja 
    ;
    Holtmann, Jorg 
    ;
    Konersmann, Marco 
    ;
    Michael, Judith 
    ;
    Neumann, Eva-Maria 
    ;
    Pfeiffer, Jerome 
    ;
    Rabiser, Rick 
    ;
    Riebisch, Matthias 
    ;
    Um den effektiven und effizienten Betrieb von Cyberphysischen Produktionssystemen (CPPSen) sicherzustellen, spielt Software eine zunehmend wichtige Rolle. Die enormen Fortschritte bei Softwareentwicklungsmethoden, welche in den letzten Jahren erzielt wurden, scheinen jedoch die aktuellen Herausforderungen der Industrie nicht zu erfüllen, weil diese die Industrie nicht oder nur langsam erreichen. In diesem Beitrag werden die Herausforderungen für die Softwareentwicklung in CPPSen aus Sicht von neun Industrievertretern aus acht europäischen Unternehmen unterschiedlicher Größe diskutiert. Um den digitalen Transformationsprozess für eine zukunftsfähige Produktion zu begleiten, wurden aus den beschriebenen Herausforderungen Perspektiven für die Forschung erarbeitet. Die Umsetzung dieser Ziele ist vor dem Hintergrund von ökonomischen, sozialen und Nachhaltigkeitsanforderungen notwendig.
  • Publication
    Metadata only
    Developing an AI-Enabled IIoT Platform - Lessons Learned from Early Use Case Validation
    (Springer International Publishing, 2023) ;
    Palmer, Gregory 
    ;
    Reimer, Svenja 
    ;
    Trong Vu, Tat 
    ;
    Do, Hieu 
    ;
    Laridi, Sofiane 
    ;
    ;
    Niederée, Claudia 
    ;
    Hildebrandt, Thomas 
    ;
    Batista, Thais
    ;
    Bures, Thomas
    ;
    Raibulet, Claudia
    ;
    Muccini, Henry
    For a broader adoption of AI in industrial production, adequate infrastructure capabilities and ecosystems are crucial. This includes easing the integration of AI with industrial devices, support for distributed deployment, monitoring, and consistent system configuration. IIoT platforms can play a major role here by providing a unified layer for the heterogeneous Industry 4.0/IIoT context. However, existing IIoT platforms still lack required capabilities to flexibly integrate reusable AI services and relevant standards such as Asset Administration Shells or OPC UA in an open, ecosystem-based manner. This is exactly what our next level Intelligent Industrial Production Ecosphere (IIP-Ecosphere) platform addresses, employing a highly configurable low-code based approach. In this paper, we introduce the design of this platform and discuss an early evaluation in terms of a demonstrator for AI-enabled visual quality inspection. This is complemented by insights and lessons learned during this early evaluation activity.
  • Publication
    Metadata only
    MLOps Challenges in Industry 4.0
    (2023)
    Leonhard Faubel 
    ;
    Klaus Schmid 
    ;
    Holger Eichelberger 
    An important part of the Industry 4.0 vision is the use of machine learning (ML) techniques to create novel capabilities and flexibility in industrial production processes. Currently, there is a strong emphasis on MLOps as an enabling collection of practices, techniques, and tools to integrate ML into industrial practice. However, while MLOps is often discussed in the context of pure software systems, Industry 4.0 systems received much less attention. So far, there is only little research focusing on MLOps for Industry 4.0. In this paper, we discuss whether MLOps in Industry 4.0 leads to significantly different challenges compared to typical Internet systems. We provide an initial analysis of MLOps approaches and identify both context-independent MLOps challenges (general challenges) as well as challenges particular to Industry 4.0 (specific challenges) and conclude that MLOps works very similarly in Industry 4.0 systems to pure software systems. This indicates that existing tools and approaches are also mostly suited for the Industry 4.0 context.
      8