The module "Intelligent User Interfaces (IUI)" covers current topics at the intersection of Human-Computer Interaction and Machine Learning. The focus is on transferring and adapting methods from the fields of Machine Learning and Artificial Intelligence to practical questions of interactive system design, always with a human-centered perspective.

Topics covered include, among others:

• Fundamentals of Artificial Intelligence and Machine Learning (including Python implementations)
• Voice-based user interfaces (Voice User Interfaces)
• Text processing and Natural Language Processing (NLP)
• Context- and environment-aware interaction in intelligent systems
• Intelligent text input systems and optimized keyboard layouts
• Recommender systems and their evaluation
• Explainability and transparency of intelligent systems (Explainable AI)
• Usable security, human–machine security, and trustworthy AI
• Introduction to human–robot interaction

After successfully completing the module, students will be able to:

• name, explain, and classify key concepts, methods, and terms in machine learning, especially neural networks, in the context of human-centered applications,
• systematically collect, prepare, and analyze data for ML systems, and critically reflect on the effects of data quality, bias, and representativeness on model behavior,
• train, configure, and optimize ML models, including selecting suitable architectures, training strategies, and hyperparameters,
• compare different model classes and learning paradigms and evaluate their suitability for specific human-centered application scenarios,
• Evaluate ML models using appropriate metrics and critically interpret results in terms of generalizability, robustness, and traceability.
• Analyze model decisions and model limitations in an understandable way and discuss their consequences for users and for use in real-world contexts.
• Implement ML-based systems as prototypes, develop them iteratively, and present and reflect on their results in a structured manner.
• Critically reflect on their own design, modeling, and evaluation decisions and derive principles for the scientifically sound, responsible design and use of machine learning systems.

Module examination: 90 minutes written examination or oral examination