005 Computerprogrammierung, Programme, Daten
Refine
Departments, institutes and facilities
- Institut für Cyber Security & Privacy (ICSP) (161)
- Institut für Verbraucherinformatik (IVI) (108)
- Fachbereich Informatik (63)
- Fachbereich Wirtschaftswissenschaften (58)
- Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE) (7)
- Fachbereich Ingenieurwissenschaften und Kommunikation (3)
- Graduierteninstitut (1)
- Institut für funktionale Gen-Analytik (IFGA) (1)
- Institute of Visual Computing (IVC) (1)
- Zentrum für Ethik und Verantwortung (ZEV) (1)
Document Type
- Conference Object (187)
- Article (76)
- Part of a Book (21)
- Book (monograph, edited volume) (12)
- Contribution to a Periodical (8)
- Working Paper (4)
- Conference Proceedings (3)
- Master's Thesis (3)
- Research Data (2)
- Doctoral Thesis (2)
Year of publication
Keywords
- Usable Security (10)
- GDPR (8)
- Cloud (5)
- HTTP (5)
- Privacy (5)
- Usable Privacy (5)
- security (5)
- usable privacy (5)
- Big Data Analysis (4)
- Global Software Engineering (4)
An essential measure of autonomy in assistive service robots is adaptivity to the various contexts of human-oriented tasks, which are subject to subtle variations in task parameters that determine optimal behaviour. In this work, we propose an apprenticeship learning approach to achieving context-aware action generalization on the task of robot-to-human object hand-over. The procedure combines learning from demonstration and reinforcement learning: a robot first imitates a demonstrator’s execution of the task and then learns contextualized variants of the demonstrated action through experience. We use dynamic movement primitives as compact motion representations, and a model-based C-REPS algorithm for learning policies that can specify hand-over position, conditioned on context variables. Policies are learned using simulated task executions, before transferring them to the robot and evaluating emergent behaviours. We additionally conduct a user study involving participants assuming different postures and receiving an object from a robot, which executes hand-overs by either imitating a demonstrated motion, or adapting its motion to hand-over positions suggested by the learned policy. The results confirm the hypothesized improvements in the robot’s perceived behaviour when it is context-aware and adaptive, and provide useful insights that can inform future developments.
Cancer is one of the leading causes of death worldwide [183], with lung tumors being the most frequent cause of cancer deaths in men as well as one of the most common cancers diagnosed in woman [40]. As symptoms often arise in advanced stages, an early diagnosis is especially important to ensure the best and earliest possible treatment. In order to achieve this, Computed Tomography (CT) scans are frequently used for tumor detection and diagnosis. We will present examples of publicly available CT image data of lung cancer patients and discuss possible methods to realize an automatic system for automated cancer diagnosis. We will also look at the recent SPIE-AAPM Lung CT Challenge [10] data set in detail and describe possible methods and challenges for image segmentation and classification based on this data set.
Airborne and spaceborne platforms are the primary data sources for large-scale forest mapping, but visual interpretation for individual species determination is labor-intensive. Hence, various studies focusing on forests have investigated the benefits of multiple sensors for automated tree species classification. However, transferable deep learning approaches for large-scale applications are still lacking. This gap motivated us to create a novel dataset for tree species classification in central Europe based on multi-sensor data from aerial, Sentinel-1 and Sentinel-2 imagery. In this paper, we introduce the TreeSatAI Benchmark Archive, which contains labels of 20 European tree species (i.e., 15 tree genera) derived from forest administration data of the federal state of Lower Saxony, Germany. We propose models and guidelines for the application of the latest machine learning techniques for the task of tree species classification with multi-label data. Finally, we provide various benchmark experiments showcasing the information which can be derived from the different sensors including artificial neural networks and tree-based machine learning methods. We found that residual neural networks (ResNet) perform sufficiently well with weighted precision scores up to 79 % only by using the RGB bands of aerial imagery. This result indicates that the spatial content present within the 0.2 m resolution data is very informative for tree species classification. With the incorporation of Sentinel-1 and Sentinel-2 imagery, performance improved marginally. However, the sole use of Sentinel-2 still allows for weighted precision scores of up to 74 % using either multi-layer perceptron (MLP) or Light Gradient Boosting Machine (LightGBM) models. Since the dataset is derived from real-world reference data, it contains high class imbalances. We found that this dataset attribute negatively affects the models' performances for many of the underrepresented classes (i.e., scarce tree species). However, the class-wise precision of the best-performing late fusion model still reached values ranging from 54 % (Acer) to 88 % (Pinus). Based on our results, we conclude that deep learning techniques using aerial imagery could considerably support forestry administration in the provision of large-scale tree species maps at a very high resolution to plan for challenges driven by global environmental change. The original dataset used in this paper is shared via Zenodo (https://doi.org/10.5281/zenodo.6598390, Schulz et al., 2022). For citation of the dataset, we refer to this article.
Sind kleinere und mittlere Unternehmen (KMU) bereits auf die Digitale Transformation vorbereitet?
(2018)
Eine von den Autoren durchgeführte Untersuchung konnte deutliche Indizien dafür ausmachen, dass viele kleinere und mittlere Unternehmen (KMU) aktuell noch keine ausreichende Reife zur Digitalen Transformation haben. Zur Lösung des Problems wird vorgeschlagen, ein agiles IT-Management-Konzept zu entwickeln, um den IT-Bereich dynamisch und ohne formalen Ballast des klassischen IT-Managements zu steuern.
A company's financial documents use tables along with text to organize the data containing key performance indicators (KPIs) (such as profit and loss) and a financial quantity linked to them. The KPI’s linked quantity in a table might not be equal to the similarly described KPI's quantity in a text. Auditors take substantial time to manually audit these financial mistakes and this process is called consistency checking. As compared to existing work, this paper attempts to automate this task with the help of transformer-based models. Furthermore, for consistency checking it is essential for the table's KPIs embeddings to encode the semantic knowledge of the KPIs and the structural knowledge of the table. Therefore, this paper proposes a pipeline that uses a tabular model to get the table's KPIs embeddings. The pipeline takes input table and text KPIs, generates their embeddings, and then checks whether these KPIs are identical. The pipeline is evaluated on the financial documents in the German language and a comparative analysis of the cell embeddings' quality from the three tabular models is also presented. From the evaluation results, the experiment that used the English-translated text and table KPIs and Tabbie model to generate table KPIs’ embeddings achieved an accuracy of 72.81% on the consistency checking task, outperforming the benchmark, and other tabular models.
In 1991 the researchers at the center for the Learning Sciences of Carnegie Mellon University were confronted with the confusing question of “where is AI” from the users, who were interacting with AI but did not realize it. Three decades of research and we are still facing the same issue with the AItechnology users. In the lack of users’ awareness and mutual understanding of AI-enabled systems between designers and users, informal theories of the users about how a system works (“Folk theories”) become inevitable but can lead to misconceptions and ineffective interactions. To shape appropriate mental models of AI-based systems, explainable AI has been suggested by AI practitioners. However, a profound understanding of the current users’ perception of AI is still missing. In this study, we introduce the term “Perceived AI” as “AI defined from the perspective of its users”. We then present our preliminary results from deep-interviews with 50 AItechnology users, which provide a framework for our future research approach towards a better understanding of PAI and users’ folk theories.
For most people, using their body to authenticate their identity is an integral part of daily life. From our fingerprints to our facial features, our physical characteristics store the information that identifies us as "us." This biometric information is becoming increasingly vital to the way we access and use technology. As more and more platform operators struggle with traffic from malicious bots on their servers, the burden of proof is on users, only this time they have to prove their very humanity and there is no court or jury to judge, but an invisible algorithmic system. In this paper, we critique the invisibilization of artificial intelligence policing. We argue that this practice obfuscates the underlying process of biometric verification. As a result, the new "invisible" tests leave no room for the user to question whether the process of questioning is even fair or ethical. We challenge this thesis by offering a juxtaposition with the science fiction imagining of the Turing test in Blade Runner to reevaluate the ethical grounds for reverse Turing tests, and we urge the research community to pursue alternative routes of bot identification that are more transparent and responsive.
AI (artificial intelligence) systems are increasingly being used in all aspects of our lives, from mundane routines to sensitive decision-making and even creative tasks. Therefore, an appropriate level of trust is required so that users know when to rely on the system and when to override it. While research has looked extensively at fostering trust in human-AI interactions, the lack of standardized procedures for human-AI trust makes it difficult to interpret results and compare across studies. As a result, the fundamental understanding of trust between humans and AI remains fragmented. This workshop invites researchers to revisit existing approaches and work toward a standardized framework for studying AI trust to answer the open questions: (1) What does trust mean between humans and AI in different contexts? (2) How can we create and convey the calibrated level of trust in interactions with AI? And (3) How can we develop a standardized framework to address new challenges?
Das Kernanliegen des Datenschutzes ist es, natürliche Personen vor nachteiligen Effekten der Speicherung und Verarbeitung der sie betreffenden Daten zu schützen. Aber viele Personen scheinen gar nicht geschützt werden zu wollen. Im Gegenteil, viele Endanwender willigen “freiwillig“ – bewusst oder unbewusst – in eine umfassende Verarbeitung ihrer personenbezogenen Daten ein. Warum tun Menschen dies? Es werden verschiedene Ursachen diskutiert (beispielsweise in [79]), hierzu gehören Uninformiertheit, mangelnde Sensibilität, das Gefühl der Hilflosigkeit, mangelnde Zahlungsbereitschaft und mangelnde Alternativen. Auch wenn dies in Einzelfällen zutrifft, so gibt es oft sehr wohl datenschutzfreundliche Alternativen. Beispielsweise existiert zu WhatsApp (als Instant Messaging App) die Alternative Threema. Threema gilt als EU-DS-GVO-konform und funktional durchaus mit WhatsApp vergleichbar [62]. Allerdings ist inzwischen die aktuelle Netzwerkgröße ein entscheidendes Auswahlkriterium: Im Januar 2018 hatte Threema 4,5 Millionen Nutzer [172], WhatsApp dagegen 1,5 Milliarden [171]. Dies ist ein Indiz dafür, dass WhatsApp sich quasi zum De-facto-Standard entwickelt hat und es für die einzelne Person nur schwer möglich ist, viele andere “zum Wechsel auf ein anderes Produkt zu bewegen. [. . . ] Bei Diensten mit Nutzerzahlen im Milliardenbereich kann von ’Freiwilligkeit’ nur noch bedingt gesprochen werden.“ [9]
Projekte des maschinellen Lernens (ML), insbesondere im Bereich der Zeitreihenanalyse, gewinnen heute zunehmend an Bedeutung. Die Bereitstellung solcher Projekte in einer Produktionsumgebung mit dem gleichen Automatisierungsgrad wie bei klassischen Softwareprojekten ist ein komplexes Unterfangen. Die Umsetzung in Produktionsumgebungen erfordert neben klassischen DevOps auch Machine Learning Operation (MLOps) Technologien und Werkzeuge. Ziel dieser Studie ist es, einen umfassenden Überblick über verfügbare MLOps Tools zu bieten und einen spezifischen Techstack für Zeitreihen ML Projekte zu entwickeln. Es werden aktuelle Trends und Werkzeuge im Bereich MLOps durch eine multivokale Literaturrecherche (MLR) untersucht und analysiert. Die Studie identifiziert passende MLOps Werkzeuge und Methoden für die Zeitreihenanalyse und präsentiert eine spezifische Implementierung einer MLOps Pipeline für die Aktienkursprognose des S&P 500. MLOps und DevOps Tools nehmen eine essenzielle Rolle bei der effektiven Konstruktion und Verwaltung von ML Pipelines ein. Bei der Auswahl geeigneter Werkzeuge ist stets eine spezifische Anpassung an die jeweiligen Projektanforderungen erforderlich. Die Bereitstellung einer detaillierten Darstellung der aktuellen MLOps Tool Landschaft erweist sich hierbei als wertvolle Ressource, die es Entwicklern ermöglicht, die Effizienz und Effektivität ihrer ML Projekte zu optimieren.
Objektrelationale Datenbanken und Rough Sets für die Analyse von Contextualized Attention Metadata
(2009)
There has been a growing interest in taste research in the HCI and CSCW communities. However, the focus is more on stimulating the senses, while the socio-cultural aspects have received less attention. However, individual taste perception is mediated through social interaction and collective negotiation and is not only dependent on physical stimulation. Therefore, we study the digital mediation of taste by drawing on ethnographic research of four online wine tastings and one self-organized event. Hence, we investigated the materials, associated meanings, competences, procedures, and engagements that shaped the performative character of tasting practices. We illustrate how the tastings are built around the taste-making process and how online contexts differ in providing a more diverse and distributed environment. We then explore the implications of our findings for the further mediation of taste as a social and democratized phenomenon through online interaction.
Herein we report an update to ACPYPE, a Python3 tool that now properly converts AMBER to GROMACS topologies for force fields that utilize nondefault and nonuniform 1–4 electrostatic and nonbonded scaling factors or negative dihedral force constants. Prior to this work, ACPYPE only converted AMBER topologies that used uniform, default 1–4 scaling factors and positive dihedral force constants. We demonstrate that the updated ACPYPE accurately transfers the GLYCAM06 force field from AMBER to GROMACS topology files, which employs non-uniform 1–4 scaling factors as well as negative dihedral force constants. Validation was performed using β-d-GlcNAc through gas-phase analysis of dihedral energy curves and probability density functions. The updated ACPYPE retains all of its original functionality, but now allows the simulation of complex glycomolecular systems in GROMACS using AMBER-originated force fields. ACPYPE is available for download at https://github.com/alanwilter/acpype.
Open-Source Software spielt sowohl zur Ausgestaltung von Lehr- und Lernszenarien (bspw. Organisation mit Editoren und Groupware, Kollaboration und Kommunikation via Chats und Webblogs), als auch für die Umsetzung von Forschunsprojekten (zum Beispiel Auswertung großer Datenbestände, Erprobung realer Situationen in vituellen Laboren, Evaluation neuer Oberflächenentwicklungen) eine wichtige Rolle. Um eine bestmögliche Passung der Software herzustellen, erfolgt Softwareentwicklung im Hochschulbereich entweder forschungsprojektbezogen oder Disziplin- und Einrichtungsübergreifend.
Software offshoring has been established as an important business strategy over the last decade. While research on such forms of Global Software Development (GSD) has mainly focused on the situation of large enterprises, small enterprises are increasingly engaging in offshoring, too. Representing the biggest share of the German software industry, small companies are known to be important innovators and market pioneers. They often regard their flexibility and customer-orientation as core competitive advantages. Unlike large corporations, their small size allows them to adopt software development approaches that are characterized by a high agility and flat hierarchies. At the same time, their distinct strategies make it unlikely that they can simply adopt management strategies that were developed for larger companies.
Flexible development approaches like the ones preferred by small corporations have proven to be problematic in the context of offshoring, as their strong dependency on constant communication is strongly affected by the various barriers of international cooperation between companies. Cooperating closely over companies’ borders in different time zones and in culturally diverse teams poses complex obstacles for flexible management approaches. It is still a matter of discussion in fields like Software Engineering and Computer Supported Cooperative Work how these obstacles can be tackled and how they affect companies in the long term. Hence, it is agreed that we need a more detailed understanding of distributed software development practices in order to come to feasible technological and organizational solutions.
This dissertation presents results from two ethnographically-informed case studies of software offshoring in small German enterprises. By adopting Anselm Strauss’ concept of articulation work, we want to deepen the understanding of managing distributed software development in flexible, customer-oriented organizations. In doing so, we show how practices of coordinating inter-organizational software development are closely related to aspects of organizational learning in small enterprises. By means of interviews with developers and project managers from both parties of the cooperation, we do not only take into account the multiple perspectives of the cooperation, but also include the socio-cultural background of international software development projects into our analysis.
An der Hochschule Bonn-Rhein-Sieg fand am Donnerstag, den 23.9.21 das erste Verbraucherforum für Verbraucherinformatik statt. Im Rahmen der Online-Tagesveranstaltung diskutierten mehr als 30 Teilnehmer:innen über Themen und Ideen rund um den Bereich Verbraucherdatenschutz. Dabei kamen sowohl Beiträge aus der Informatik, den Verbraucher- und Sozialwissenschaften sowie auch der regulatorischen Perspektive zur Sprache. Der folgende Beitrag stellt den Hintergrund der Veranstaltung dar und berichtet über Inhalte der Vorträge sowie Anknüpfungspunkte für die weitere Konstituierung der Verbraucherinformatik. Veranstalter waren das Institut für Verbraucherinformatik an der H-BRS in Zusammenarbeit mit dem Lehrstuhl IT-Sicherheit der Universität Siegen sowie dem Kompetenzzentrum Verbraucherforschung NRW der Verbraucherzentrale NRW e. V. mit Förderung des Bundesministeriums der Justiz und für Verbraucherschutz.