Listen
38 Ergebnisse
Bereich: Suchergebnisse
Publikation A flexible, extendable and adaptable model to support AI coaching(Springer, 2023) Duhan, Ritu; Pande, Charuta; Martin, Andreas; Hinkelmann, Knut; López-Pellicer, Francisco J.; Polini, AndreaWe present a model based on coaching definitions, concepts, and theories to support AI coaching. The model represents the evidence-based coaching practice in different coaching domains by identifying the common elements in the coaching process. We then map the elements of the coaching model with Conversational AI design and development strategies to highlight how an AI coach can be instantiated from the model. We showcase the instantiation through an example use case of an HIV coaching chatbot.04B - Beitrag KonferenzschriftPublikation Evaluation of synthetic data generators on complex tabular data(Springer, 2024) Thees, Oscar; Novak, Jiri; Templ, Matthias; Domingo-Ferrer, Josep; Önen, MelekSynthetic data generators are widely utilized to produce synthetic data, serving as a complement or replacement for real data. However, the utility of data is often limited by its complexity. The aim of this paper is to show their performance using a complex data set that includes cluster structures and complex relationships. We compare different synthesizers such as synthpop, Synthetic Data Vault, simPop, Mostly AI, Gretel, Realtabformer, and arf, taking into account their different methodologies with (mostly) default settings, on two properties: syntactical accuracy and statistical accuracy. As a complex and popular data set, we used the European Statistics on Income and Living Conditions data set. Almost all synthesizers resulted in low data utility and low syntactical accuracy. The results indicated that for such complex data, simPop, a computational and methodological framework for simulating complex data based on conditional modeling, emerged as the most effective approach for static tabular data and is superior compared to other conditional or joint modelling approaches.04B - Beitrag KonferenzschriftPublikation How to support customer segmentation with useful cluster descriptions(Springer, 2015) Witschel, Hans Friedrich; Loo, Simon; Riesen, Kaspar; Perner, PetraCustomer or market segmentation is an important instrument for the optimisation of marketing strategies and product portfolios. Clustering is a popular data mining technique used to support such segmentation – it groups customers into segments that share certain demographic or behavioural characteristics. In this research, we explore several automatic approaches which support an important task that starts after the actual clustering, namely capturing and labeling the “essence” of segments. We conducted an empirical study by implementing several of these approaches, applying them to a data set of customer representations and studying the way our study participants interacted with the resulting cluster representations. Major goal of the present paper is to find out which approaches exhibit the greatest ease of understanding on the one hand and which of them lead to the most correct interpretation of cluster essence on the other hand. Our results indicate that using a learned decision tree model as a cluster representation provides both good ease of understanding and correctness of drawn conclusions.04B - Beitrag KonferenzschriftPublikation Challenges, opportunities and application fields of quantum computing - an introductory overview(Association Information et Management, 2022) Kech, Benjamin; Schneider, Bettina; Gachnang, Phillip; Azan, WilfridThis paper elaborates on the technology of quantum computing. It is aimed at people new to this field and introduces characteristics of quantum computing along with comparisons to classical computing. Furthermore, the paper describes the challenges and opportunities of quantum computing. In addition, applications have been explored where quantum computing could create value for businesses. The findings include three categories. First, quantum computing might make a difference in simulating nature, which was also the initial idea that led to the invention of quantum computing. Second, quantum computing could benefit the category of machine learning. Last, optimization problems will take advantage of quantum computing. It is concluded that quantum computing is still in its early stages and there are many challenges to overcome ‚ in particular the challenge of error correction. To gain in the foreseeable future from the advantages that quantum computing pledges, more advances in research have to be made to have a fault-tolerant system. A fault-tolerant quantum computer is a promising technology that could create significant value for various branches, such as finance.04B - Beitrag KonferenzschriftPublikation Was man mit Blockchain nicht machen kann(Netzmedien, 19.06.2019) Dettling, WalterNach der ersten Begeisterung über Kryptowährungen und Blockchain ist eine gewisse Ernüchterung eingetreten. Im Gegensatz zu Kryptowährungen ist bei den übrigen Blockchain-Anwendungen ist noch keine «Killerapplikation» in Sicht. Grund dafür sind die völlig unterschiedliche Konzeptionen von herkömmlichen Datenstrukturen für Business Applikationen und dem Konzept der Blockchain, welche primär den Schutz der Daten vor Manipulation und die Anonymität gewährleisten.01B - Beitrag in Magazin oder ZeitungPublikation Shaping aggregated load profiles based on optimized local scheduling of home appliances(Springer, 01.02.2018) Hunziker, Christoph; Schulz, Nicola; Wache, HolgerWe present a new method to control an aggregated electric load profile by exploiting the flexibilities provided by residential homes. The method is based on a common energy price combined with inclining block rates, broadcasted to all households allowing them to minimize their energy provisioning cost. The distributed home energy management systems receive the price signal and use mixed integer linear programming for optimal scheduling of load, storage, and generation devices. The method provides excellent scalability as well as autonomy for home owners and avoids load synchronization effects. As proof of concept, an optimization algorithm for determining a day-ahead price is applied in two case studies. An excellent conformance between a given reference load profile and the resulting aggregated load profile of all households is demonstrated.01A - Beitrag in wissenschaftlicher ZeitschriftPublikation A market-based smart grid approach to increasing power grid capacity without physical grid expansion(Springer, 01.02.2018) Bagemihl, Joachim; Boesner, Frank; Riesinger, Jens; Künzli, Michael; Wilke, Gwendolin; Binder, Gabriela; Wache, Holger; Laager, Daniel; Breit, Jürgen; Wurzinger, Michael; Zapata, Juliana; Ulli-Beer, Silvia; Layec, Vincent; Stadler, Thomas; Stabauer, FranzThe continuous increase of competitiveness of renewable energy in combination with the necessity of fossil fuel substitution leads to further electrification of the global energy system and therefore a need for large-scale power grid capacity increase. While physical grid expansion is not feasible for many countries, grid-driven energy management in the Smart Grid often interferes in customer processes and free access to the energy market. The paper solves this dilemma by proposing a market-based load schedule management approach that increases power grid capacity without physical grid expansion. This is achieved by allocating for a certain class of non-critical flexible loads called “conditional loads” the currently unused grid capacity dedicated to ensuring N−1 security of supply whereas this security level remains untouched for all critical processes. The paper discusses the necessary processes and technical and operational requirements to operate such a system.01A - Beitrag in wissenschaftlicher ZeitschriftPublikation Technical Validation of the RLS Smart Grid Approach to increase Power Grid Capacity without Physical Grid Expansion(SciTePress, 05/2019) Christen, Ramón; Layec, Vincent; Wilke, Gwendolin; Wache, Holger; Donnellan, Brian; Klein, Cornel; Helfert, MarkusThe electrification of the global energy system and the shift towards distributed power production from sus- tainable sources triggers an increased network capacity demand at times of high production or consumption. Existing energy management solutions can help mitigate resulting high costs of large-scale physical grid rein- forcement, but often interfere in customer processes or restrict free access to the energy market. In a preceding paper, we proposed the RLS regional load shaping approach as a novel business model and load management solution in middle voltage grid to resolve this dilemma: market-based incentives for all stakeholders are pro- vided to allow for flexible loads that are non-critical in customer processes to be allocated to the unused grid capacity traditionally reserved for N-1 security of supply. We provide a validation of the technical aspects of the approach, with an evaluation of the day-ahead load forecasting method for industry customers and a load optimization heuristics. The latter is tested by a simulation run on a scenario of network branch with provoked capacity bottlenecks. The method handles all provoked critical network capacity situations as expected.04B - Beitrag KonferenzschriftPublikation Poster abstract: SmartStability. A multi-agent simulation environment for flexibility trading in households(Springer, 01.02.2018) Wache, Holger; Künzli, Michael; Schulz, Nicola; Bichsel, Jürg; Hall, MonikaThe increasing number of volatile energy sources, such as solar power plants, challenges the power network operators, the energy brokers as well as the electricity market actors. In this work, a multi-agent based approach will be introduced that allows multiple households to trade flexibilities, on top of the usual selling of produced energy to the paying consumer. Flexibility trading allows different optimisations for different actors. They can slightly shift their electricity consumption, e.g. by turning boilers on/off, to optimise the system (e.g. follow a predefined schedule). Simulation results indicate that a flexibility market consisting of only few households can already help to optimise the system.01A - Beitrag in wissenschaftlicher ZeitschriftPublikation Event Data Sources for Enterprise Architecture Documentation(Hochschule für Wirtschaft FHNW, 02/2015) Frei, Michael; Hinkelmann, KnutEnterprise Architecture Management (EAM) has become a widely accepted discipline that concerns itself with the alignment of business and information technology (IT) and deals with the increasing complexity of information system (IS) landscapes. It is the practice of documenting the interrelations of business and IT entities by developing a model of the enterprise architecture (EA). This documentation of the EA acts as basis for planning the future architecture and prioritizing the projects that transform the current architecture. Due to globalizing markets and the fast paced evolution of IT, organizations are forced to become agile and frequently adapt their internal architecture in order to stay competitive. Today, the majority of tasks concerning the maintenance of EA documentation are based on manual methods and are connected to huge data collection efforts. This has a negative impact on the quality and reliability of the EA documentation and therefore diminishes the benefits of EAM. The popularly utilized EA frameworks cover the process of keeping the EA documentation up-to-date only marginally. In recent literature, there has been an effort to automate the EA documentation with the use and integration of different productive IS across the company. It was identified that common IS in organizations not only hold data about the EA, but also can contain events that could be used to trigger EA documentation processes. However, only few concrete implementations of automated EA documentation are examined in research or found in practice, because organizations face several challenges when attempting to integrate IS into an automated EA documentation. This thesis identifies possible sources of EA data and EA events in the form of common IS or databases in today’s enterprises. Further, the thesis develops an evaluation catalogue for potential source candidates in order to recommend a set of suitable documentation techniques for each source. A proof of concept through a case-study at a financial institution demonstrates the feasibility and applicability of the catalogue. The proposed data source selection catalogue for EA documentation supports the identification and selection of productive IS and suitable techniques in order to increase the quality of EA documentation11 - Studentische Arbeit