Now showing 1 - 10 of 30
  • Publication
    Metadata only
    Behavioral Avoidance Test: Comparison between in vivo and virtual reality using questionnaires and psychophysiology
    (IEEE, 2022-12-14)
    Schmuecker, Vanessa
    ;
    ;
    Hildebrand, Anne Sophie
    ;
    Jakob, Rebekka
    ;
    Eiler, Tanja Joan
    ;
    ;
    Klucken, Tim
    ;
    Brueck, Rainer
    The field of virtual reality (VR) is no longer limited to the entertainment sector, and the use also increased in the sector of healthcare and therapy. This paper attempts to compare the effectiveness of a Behavioral Avoidance Test for spider phobia in virtual reality to in vivo. To do this, two different BAT approaches are designed for both in vivo and in virtuo to be compared by a group of phobic participants. In addition, physiological data of the participants were recorded to investigate this comparison on a physiological level.
  • Publication
    Metadata only
    Demographic Aware Probabilistic Medical Knowledge Graph Embeddings of Electronic Medical Records
    (Springer, 2021-03-22)
    Guluzade, Aynur
    ;
    Kacupaj, Endri
    ;
    Medical knowledge graphs (KGs) constructed from Electronic Medical Records (EMR) contain abundant information about patients and medical entities. The utilization of KG embedding models on these data has proven to be efficient for different medical tasks. However, existing models do not properly incorporate patient demographics and most of them ignore the probabilistic features of the medical KG. In this paper, we propose DARLING (Demographic Aware pRobabiListic medIcal kNowledge embeddinG), a demographic-aware medical KG embedding framework that explicitly incorporates demographics in the medical entities space by associating patient demographics with a corresponding hyperplane. Our framework leverages the probabilistic features within the medical entities for learning their representations through demographic guidance. We evaluate DARLING through link prediction for treatments and medicines, on a medical KG constructed from EMR data, and illustrate its superior performance compared to existing KG embedding models.
  • Publication
    Metadata only
    Conversational question answering over knowledge graphs with transformer and graph attention networks
    (Association for Computational Linguistics (ACL), 2021-01-01)
    Kacupaj, Endri
    ;
    Plepi, Joan
    ;
    Singh, Kuldeep
    ;
    Thakkar, Harsh
    ;
    Lehmann, Jens
    ;
    This paper addresses the task of (complex) conversational question answering over a knowledge graph. For this task, we propose LASAGNE (muLti-task semAntic parSing with trAnsformer and Graph atteNtion nEtworks). It is the first approach, which employs a transformer architecture extended with Graph Attention Networks for multi-task neural semantic parsing. LASAGNE uses a transformer model for generating the base logical forms, while the Graph Attention model is used to exploit correlations between (entity) types and predicates to produce node representations. LASAGNE also includes a novel entity recognition module which detects, links, and ranks all relevant entities in the question context. We evaluate LASAGNE on a standard dataset for complex sequential question answering, on which it outperforms existing baseline averages on all question types. Specifically, we show that LASAGNE improves the F1-score on eight out of ten question types; in some cases, the increase in F1-score is more than 20% compared to the state of the art.
  • Publication
    Metadata only
    MLM: A Benchmark Dataset for Multitask Learning with Multiple Languages and Modalities
    (Association for Computing Machinery, 2020-10-19)
    Armitage, Jason
    ;
    Kacupaj, Endri
    ;
    Tahmasebzadeh, Golsa
    ;
    Swati
    ;
    ;
    Ewerth, Ralph
    ;
    Lehmann, Jens
    In this paper, we introduce the MLM (Multiple Languages and Modalities) dataset - a new resource to train and evaluate multitask systems on samples in multiple modalities and three languages. The generation process and inclusion of semantic data provide a resource that further tests the ability for multitask systems to learn relationships between entities. The dataset is designed for researchers and developers who build applications that perform multiple tasks on data encountered on the web and in digital archives. A second version of MLM provides a geo-representative subset of the data with weighted samples for countries of the European Union. We demonstrate the value of the resource in developing novel applications in the digital humanities with a motivating use case and specify a benchmark set of tasks to retrieve modalities and locate entities in the dataset. Evaluation of baseline multitask and single task systems on the full and geo-representative versions of MLM demonstrate the challenges of generalising on diverse data. In addition to the digital humanities, we expect the resource to contribute to research in multimodal representation learning, location estimation, and scene understanding.
  • Publication
    Metadata only
    SOLIOT-Decentralized data control and interactions for IoT
    (MDPI, 2020-06-01)
    Bader, Sebastian R.
    ;
    The digital revolution affects every aspect of society and economy. In particular, the manufacturing industry faces a new age of production processes and connected collaboration. The underlying ideas and concepts, often also framed as a new "Internet of Things", transfer IT technologies to the shop floor, entailing major challenges regarding the heterogeneity of the domain. On the other hand, web technologies have already proven their value in distributed settings. SOLID (derived from "social linked data") is a recent approach to decentralize data control and standardize interactions for social applications in the web. Extending this approach towards industrial applications has the potential to bridge the gap between theWorldWideWeb and local manufacturing environments. This paper proposes SOLIOT-a combination of lightweight industrial protocols with the integration and data control provided by SOLID. An in-depth requirement analysis examines the potential but also current limitations of the approach. The conceptual capabilities are outlined, compared and extended for the IoT protocols CoAP and MQTT. The feasibility of the approach is illustrated through an open-source implementation, which is evaluated in a virtual test bed and a detailed analysis of the proposed components.
  • Publication
    Metadata only
    VQuAnDa: Verbalization QUestion ANswering DAtaset
    (Springer, 2020-05-27)
    Kacupaj, Endri
    ;
    Zafar, Hamid
    ;
    Lehmann, Jens
    ;
    Question Answering (QA) systems over Knowledge Graphs (KGs) aim to provide a concise answer to a given natural language question. Despite the significant evolution of QA methods over the past years, there are still some core lines of work, which are lagging behind. This is especially true for methods and datasets that support the verbalization of answers in natural language. Specifically, to the best of our knowledge, none of the existing Question Answering datasets provide any verbalization data for the question-query pairs. Hence, we aim to fill this gap by providing the first QA dataset VQuAnDa that includes the verbalization of each answer. We base VQuAnDa on a commonly used large-scale QA dataset – LC-QuAD, in order to support compatibility and continuity of previous work. We complement the dataset with baseline scores for measuring future training and evaluation work, by using a set of standard sequence to sequence models and sharing the results of the experiments. This resource empowers researchers to train and evaluate a variety of models to generate answer verbalizations.
  • Publication
    Metadata only
    Training multimodal systems for classification with multiple objectives
    (2020-01-01)
    Armitage, Jason
    ;
    Thakur, Shramana
    ;
    Tripathi, Rishi
    ;
    Lehmann, Jens
    ;
    We learn about the world from a diverse range of sensory information. Automated systems lack this ability as investigation has centred on processing information presented in a single form. Adapting architectures to learn from multiple modalities creates the potential to learn rich representations of the world - but current multimodal systems only deliver marginal improvements on unimodal approaches. Neural networks learn sampling noise during training with the result that performance on unseen data is degraded. This research introduces a second objective over the multimodal fusion process learned with variational inference. Regularisation methods are implemented in the inner training loop to control variance and the modular structure stabilises performance as additional neurons are added to layers. This framework is evaluated on a multilabel classification task with textual and visual inputs to demonstrate the potential for multiple objectives and probabilistic methods to lower variance and improve generalisation.
  • Publication
    Metadata only
    The Semantic Asset Administration Shell
    (Springer, 2019-11)
    Bader, Sebastian R.
    ;
    The disruptive potential of the upcoming digital transformations for the industrial manufacturing domain have led to several reference frameworks and numerous standardization approaches. On the other hand, the Semantic Web community has made significant contributions in the field, for instance on data and service description, integration of heterogeneous sources and devices, and AI techniques in distributed systems. These two streams of work are, however, mostly unrelated and only briefly regard the each others requirements, practices and terminology. We contribute to this gap by providing the Semantic Asset Administration Shell, an RDF-based representation of the Industrie 4.0 Component. We provide an ontology for the latest data model specification, created a RML mapping, supply resources to validate the RDF entities and introduce basic reasoning on the Asset Administration Shell data model. Furthermore, we discuss the different assumptions and presentation patterns, and analyze the implications of a semantic representation on the original data. We evaluate the thereby created overheads, and conclude that the semantic lifting is manageable, also for restricted or embedded devices, and therefore meets the conditions of Industrie 4.0 scenarios.
  • Publication
    Metadata only
    Structuring reference architectures for the industrial Internet of Things
    (Basel, 2019-07-01)
    Bader, Sebastian R.
    ;
    ;
    Lohmann, Steffen
    The ongoing digital transformation has the potential to revolutionize nearly all industrial manufacturing processes. However, its concrete requirements and implications are still not sufficiently investigated. In order to establish a common understanding, a multitude of initiatives have published guidelines, reference frameworks and specifications, all intending to promote their particular interpretation of the Industrial Internet of Things (IIoT). As a result of the inconsistent use of terminology, heterogeneous structures and proposed processes, an opaque landscape has been created. The consequence is that both new users and experienced experts can hardly manage to get an overview of the amount of information and publications, and make decisions on what is best to use and to adopt. This work contributes to the state of the art by providing a structured analysis of existing reference frameworks, their classifications and the concerns they target. We supply alignments of shared concepts, identify gaps and give a structured mapping of regarded concerns at each part of the respective reference architectures. Furthermore, the linking of relevant industry standards and technologies to the architectures allows a more effective search for specifications and guidelines and supports the direct technology adoption.
  • Publication
    Metadata only
    DLUBM: A benchmark for distributed linked data knowledge base systems
    (Springer, 2017-10-21)
    Keppmann, Felix Leif
    ;
    ;
    Harth, Andreas
    Linked Data is becoming a stable technology alternative and is no longer only an innovation trend. More and more companies are looking into adapting Linked Data as part of the new data economy. Driven by the growing availability of data sources, solutions are constantly being newly developed or improved in order to support the necessity for data exchange both in web and enterprise settings. Unfortunately, currently the choice whether to use Linked Data is more an educated guess than a fact-based decision. Therefore, the provisioning of open benchmarking tools and reports, which allow developers to assess the fitness of existing solutions, is key for pushing the development of better Linked Data-based approaches and solutions. To this end we introduce a novel Linked Data benchmark – Distributed LUBM, which enables the reproducible creation and deployment of distributed interlinked LUBM datasets. We provide a system architecture for distributed Linked Data benchmark environments, accompanied by guiding design requirements. We instantiate the architecture with the actual DLUBM implementation and evaluate a Linked Data query engine via DLUBM.