Theses

In the Internet of Things (IoT) landscape, large networks of sensors collect data that drive crucial insights and decisions. However, the reliability of this data can be compromised due to environmental noise, sensor drift, or hardware inconsistencies. Sensor data fusion—the process of combining data from multiple sensors—can help overcome these challenges by producing a more accurate, consistent, and reliable estimation of the monitored system. The objective of this thesis is to develop a robust mechanism to fuse sensor data, focusing on techniques that validate and enhance data accuracy. By applying sensor fusion techniques, the aim is to design a system that improves data reliability in dynamic IoT environments, creating the basis for smarter, data-driven systems.

Pedagogical Conversational Agents (PCAs) are increasingly utilized in educational settings to facilitate and enhance the learning process. These agents are commonly implemented as text-based chatbots, functioning as conversational interfaces that provide instructional support. PCAs can serve as tutors or motivators, assisting learners in achieving their educational objectives. This thesis aims to integrate anthropomorphic design elements, i.e., features that enhance the agent’s human-like appearance and interaction, into an existing tutoring chatbot. The primary objectives of this research are to conduct a comprehensive review of relevant literature, implement human-like design features, and evaluate their impact on user experience and learning outcomes.

This thesis develops comprehensive assessment mechanisms for evaluating Generative AI-based Internet of Things applications. Evaluating artifacts powered by nondeterministic AI algorithms presents inherent complexity, requiring sophisticated frameworks that address both technical performance and practical effectiveness. The research establishes multi-dimensional evaluation criteria encompassing accuracy, latency, scalability, user satisfaction, and interpretability. These metrics are benchmarked against established standards while adapting recent frameworks for assessing large language models to address the unique temporal and contextual characteristics of IoT data streams. Through systematic review of existing assessment methodologies and analysis of current GenAI-IoT applications, this work proposes novel evaluation approaches tailored to this emerging field. The outcome includes evaluation methodologies, practical guidelines, and tools designed to measure artifact performance in Design Science Research contexts. These assessment mechanisms support iterative refinement cycles and provide structured approaches for validating GenAI-IoT integration effectiveness across diverse deployment scenarios.

This thesis conceptualizes an interaction model that describes the connection between IoT applications and generative AI in a detailed and structured manner. Focusing mainly on the technological and partly on the organizational dimension, the model follows an IoT data pipeline comprising data integration, processing in data platforms, and use in applications, and systematically captures the interactions of generative AI at each stage. The work follows a Design Science Research approach, builds on a structured qualitative literature review, develops a technically grounded conceptual interaction model and framework, and reflects the result in the context of the existing IoT research agenda.

Based on a systematic literature review, this bachelor’s thesis identifies use cases described in the scientific literature for the integration of Internet of Things (IoT) systems and Large Language Models (LLMs). It examines scenarios in which LLMs analyze, interpret, or generate IoT-related data in domains such as smart cities, enterprises, smart homes, Industry 4.0, individual use, and healthcare. The identified application contexts are categorized and assessed with regard to opportunities, technical and data-related challenges such as large data volumes, high data rates, heterogeneous sources, and temporal or contextual dependencies, as well as risks and ethical questions including privacy, surveillance, and fairness.

This bachelor’s thesis conceives, implements, and evaluates an LLM-supported system for linking complementary IoT data sources as a functional IT artifact. The system is intended to enable ad-hoc analyses in a university campus environment without requiring users to have technical or statistical expertise. It will make multiple SensorThings-API-based data sources accessible to a Large Language Model in a standardized and context-aware way, dynamically integrate data such as library and room occupancy, CO₂ levels, and temperature, and coordinate cross-source analyses through tool calls. The artifact follows Design Science Research Methodology and is evaluated for user acceptance, correctness of data-source selection, and reliability of the generated analyses.

This master’s thesis designs and evaluates a prototypical LLM-based agent system for selected tasks of personal financial advisors in order to assess the task-technology fit of such systems and derive design principles for agentic financial advisory systems. It identifies and structures advisory work activities, translates task-specific requirements into agent capabilities, evaluates existing financial-advisor agent solutions, implements a prototype for a defined subset of tasks, and evaluates it with regard to task-technology fit, task performance, reliability, perceived usefulness, and user acceptance. The research follows Design Science Research Methodology and combines literature-based task analysis, prototyping, scenario-based and benchmark-informed tests, and semi-structured interviews with potential users or domain experts.