Research Group on Human-Computer Interaction & Visual Analytics (VIS)

About the lab

The research group of Visual Analytics and Human-Computer-Interaction of the Darmstadt University of Applied Sciences (h_da | vis) researches and works on solutions on the basis of Big Data with the main focus on industrial benefits. Visual Analytics together with Data Enrichment and Data Mining is a key enabler to gather insights and finally helps in performing better decision-making.

Our major work is based on our novel Scitics technology and is aligned to Business Analytics, where we aim to support analysts in analyzing and extracting trends by using public, research and internal data for technology, innovation and strategy management.

More information you’ll find under:

Featured projects (4)

The book "ICTE in Transportation and Logistics" is interdisciplinary annual issue published by Springer Nature Switzerland AG on the edge between transportation, logistics, economy and computer science highlighting sociotechnical aspects of any real sustainable system. The issue would be the announcing area of successful research projects giving possibilities for fast dissemination the information about new findings. The book will be covered by Scopus and Web of Science.
The project “AVARTIM” is to be used to develop a software-supported process for recognizing and evaluating trends, market and technology signals in order to sustainably support the process of innovation and technology management. As part of the project, an infrastructure will be set up at Darmstadt University of Applied Sciences, which is modular and thus able to react quickly to technological changes. The infrastructure to be developed here serves as preliminary research and initial technology both for industrial use by and with the SME partners as well as for the application for joint projects. First of all, participation in the LOEWE funding line 3 of the state of Hessen and subsequently in the tender for LEIT-ICT / Big Data technologies of the EU is aimed at. More information on:
The main goal of this project is to create a sustainable European strategic alliance to foster European research and innovation in the area of Visual Analytics, Artificial Intelligence, Simulation, Prediction and Planning of emerging technologies and innovations for business and eGovernance. The strategic alliance is set up for a long period and serves the purpose to discover continuously new opportunities to strengthen the European activities of the Darmstadt University of Applied Sciences. It aims at investigating various European funding and networking opportunities and submitting proposals to the various European research‐related programs. This project targets networking activities to help setting up the strategic alliance. More information on:

Featured research (29)

The automatic detection of disinformation and misinformation has gained attention during the last years, since fake news has a critical impact on democracy, society, and journalism and digital literacy. In this paper, we present a binary content-based classification approach for detecting fake news automatically, with several recently published pre-trained language models based on the Transformer architecture. The experiments were conducted on the FakeNewsNet dataset with XLNet, BERT, RoBERTa, DistilBERT, and ALBERT and various combinations of hyperparameters. Different preprocessing steps were carried out with only using the body text, the titles and a concatenation of both. It is concluded that Transformers are a promising approach to detect fake news, since they achieve notable results, even without using a large dataset. Our main contribution is the enhancement of fake news’ detection accuracy through different models and parametrizations with a reproducible result examination through the conducted experiments. The evaluation shows that already short texts are enough to attain 85% accuracy on the test set. Using the body text and a concatenation of both reach up to 87% accuracy. Lastly, we show that various preprocessing steps, such as removing outliers, do not have a significant impact on the models prediction output.
Die visuelle Projektion von heterogenen (z.B. Forschungs-)Daten auf einer 2-dimensionalen Fläche, wie etwa einem Bildschirm, wird als Datenvisualisierung bezeichnet. Datenvisualisierung ist ein Oberbegriff für verschiedene Arten der visuellen Projektion. In diesem Kapitel wird zunächst der Begriff definiert und abgegrenzt. Der Fokus des Kapitels liegt auf Informationsvisualisierung und Visual Analytics. In diesem Kontext wird der Prozess der visuellen Transformation vorgestellt. Es soll als Grundlage für eine wissenschaftlich valide Generierung von Visualisierungen dienen, die auch visuelle Aufgaben umfassen. Anwendungsszenarien stellen den Mehrwert der hier vorgestellten Konzepte in der Praxis vor. Der wissenschaftliche Beitrag liegt in einer formalen Definition des visuellen Mappings.
Cyber-physical systems in smart factories get more and more integrated and interconnected. Industry 4.0 accelerates this trend even further. Through the broad interconnectivity a new class of faults arise, the contextual faults, where contextual knowledge is needed to find the underlying reason. Fully-automated systems and the production line in a smart factory form a complex environment making the fault diagnosis non-trivial. Along with a dataset, we give a first definition of contextual faults in the smart factory and name initial use cases. Additionally, the dataset encompasses all the data recorded in a current state-of-the-art smart factory. We also add additional information measured by our developed sensing units to enrich the smart factory data even further. In the end, we show a first approach to detect the contextual faults in a manual preliminary analysis of the recorded log data.
A variety of new technologies and ideas for businesses are arising in the domain of logistics and mobility. It can be differentiated between fundamental new approaches, e.g. central packaging stations or deliveries via drones and minor technological advancements that aim on more ecologically and economic transportation. The need for analytical systems that enable identifying new technologies, innovations, business models etc. and give also the opportunity to rate those in perspective of business relevance is growing. The users’ behavior is commonly investigated in adaptive systems, which is considering the induvial preferences of users, but neglecting often the tasks and goals of the analysis. A process-related supports could assist to solve an analytical task in a more efficient and effective way. We introduce in this paper an approach that enables non-professionals to perform visual trend analysis through an advanced process assistance based on process mining and visual adaptation. This allows generating a process model based on events, which is the baseline for process support feature calculation. These features in form of visual adaptations and the process model enable assisting non-experts in complex analytical tasks.

Lab head

Kawa Nazemi
  • Faculty of Media
About Kawa Nazemi
  • I am a full professor for Human-Computer Interaction & Visual Analytics at the Darmstadt University of Applied Sciences. Me and my research group investigate in particular machine learning methods combined with humans' perception and cognition to enable new insights from heterogenous data. I am further adjunct senior lecturer at Cork Institute of Technology and adjunct lecturer at the Technische Universität Darmstadt.

Members (5)

Dirk Burkhardt
  • Software AG
Nicola Below
  • Darmstadt University of Applied Sciences
Lukas Kaupp
  • Darmstadt University of Applied Sciences
Lennart B. Sina
  • Darmstadt University of Applied Sciences
Midhad Blazevic
  • Darmstadt University of Applied Sciences