Search
Generic filters
Exact matches only
Filter by content type
Users
Attachments

Hybride Intelligenz

Übersicht

Wie können menschliche Eigenschaften wie Erfahrung, Kreativität oder Intuition mit Machine Learning und künstlicher Intelligenz kombiniert werden, so dass die unterschiedlichen Stärken bestmöglich kombiniert werden? Zur Beantwortung dieser Frage versuchen wir datenbasierete Entscheidungs-, Innovations- und Arbeitsprozesse intelligent zu verbessern und / oder zu automatisieren.


Schwerpunkte

  • Datenbasierte Entscheidungsprozesse
  • Unterstützung von Management-Entscheidungen mittels Data Science und Predictive Analytics
  • Gestaltung von intelligenten Entscheidungsunterstützungssystemen
  • Arbeitsteilung zwischen Menschlicher und Künstlicher Intelligenz (Human vs. AI-in-the Loop)
  • Kombination kollektiver und künstlicher Intelligenz
  • Analyse grosser Mengen unstrukturierter Daten für die Entscheidungsunterstützung
  • Explainable AI und Design von Explanation Interfaces

Projekte

Hybrid Creativity: How Artificial Intelligence Can Defy and Reinforce Barriers in Human Creative Performance

A highly successful strategy to thrive innovation in private and public organizations is the integration of consumers and external stakeholders into the innovation process. Eighty five percent of the 100 most valued brands already use digital platforms such as OpenIdea, Crowdspring, HyveCrowd or propretiary platforms such as LEGO ideas.

Kontakt
Mehr

Verbindung künstlicher und kollektiver Intelligenz zur Entwicklung skalierbarer Software Testing Lösungen

"Software is eating the world" - Aber wie stellt man sicher, dass diese Software auch benutzerfreundlich ist und richtig funktioniert? Ziel des Projekts ist es, selbstlernende Textanalyse-Algorithmen zu entwickeln und zu untersuchen, inwieweit diese künstliche Intelligenz beim Softwaretest in Unternehmen wirtschaftlich vorteilhaft und sinnvoll eingesetzt werden kann.

Kontakt
Mehr

Publikationen

Investors increasingly use machine learning (ML) algorithms to support their early stage investment decisions. However, it remains unclear if algorithms can make better investment decisions and if so, why. Building on behavioral decision theory, our study compares the investment returns of an algorithm with those of 255 business angels (BAs) investing via an angel investment platform. We explore the influence of human biases and experience on BAs’ returns and find that investors only outperformed the algorithm when they had extensive investment experience and managed to suppress their cognitive biases. These results offer novel insights into the role of cognitive limitations, experience, and the use of algorithms in early stage investing.

Mehr
Crowdsourcing represents a powerful approach for organizations to collect data from large networks of people. While research already made great strides to develop the technological foundations for processing crowdsourced data, little is known about decision-making patterns that emerge when decision-makers have access to such large amounts of data on people's behavior, opinions, or ideas. In this study, we analyze the characteristics of decision-making in crowdsourcing based on interviews with decision-makers across 10 multinational corporations. For research, we identify four common patterns of decision-making that range from structured and goal-oriented to highly dynamic and data-driven. In this way, we systematize how decision-makers typically source, process, and use crowdsourced data to inform decisions. We also provide an integrated perspective on how different types of decision problems and modes of acquiring information induce such patterns. For practice, we discuss how information systems should be designed to provide adequate support for these patterns.

Mehr
Crowdsourcing represents a powerful approach that seeks to harness the collective knowledge or creativity of a large and independent network of people for organizations. While the approach drastically facilitates the sourcing and aggregating of information, it represents a latent challenge for organizations to process and evaluate the vast amount of crowdsourced contributions – especially when they are submitted in an unstructured, textual format. In this study, we present an on-going design science research project that is concerned with the construction of a design theory for semi-automated information processing and decision support in crowdsourcing. The proposed concept leverages the power of crowdsourcing in combination with text mining and machine learning algorithms to make the evaluation of textual contributions more efficient and effective for decision-makers. Our work aims to provide the theoretical foundation for designing such systems in crowdsourcing. It is intended to contribute to decision support and business analytics research by outlining the capabilities of text mining and machine learning techniques in contexts that face large amounts of user-generated content. For practitioners, we provide a set of generalized design principles and design features for the implementation of these algorithms on crowdsourcing platforms.

Mehr