Martin R. Fischer

Hesam Hamledari, Martin Fischer

Supply chain integration remains an elusive goal for the construction and engineering industry. The high degree of fragmentation and the reliance on third-party financial institutions has pushed the physical and financial supply chains apart. The paper demonstrates how blockchain-based crypto assets (crypto currencies and crypto tokens) can address this limitation when used for conditioning the flow of funds based on the flow of products. The paper contrasts the integration between cash and product flows in supply chains that rely on fiat currencies and crypto assets for their payment settlement. Two facets of crypto asset-enabled integration, atomicity and granularity, are further introduced. The thesis is validated in the context of construction progress payments. The as-built data captured by unmanned aerial and ground vehicles was passed to an autonomous smart contract-based method that utilizes crypto-currencies and crypto tokens for payment settlement; the resulting payment datasets, written to the Ethereum blockchain, were analyzed in terms of their integration of product and cash flow. The work is concluded with a discussion of findings and their implications for the industry.

Jonas Pfeiffer, Christian M. Meyer, Claudia Schulz et al.

Our proposed system FAMULUS helps students learn to diagnose based on automatic feedback in virtual patient simulations, and it supports instructors in labeling training data. Diagnosing is an exceptionally difficult skill to obtain but vital for many different professions (e.g., medical doctors, teachers). Previous case simulation systems are limited to multiple-choice questions and thus cannot give constructive individualized feedback on a student's diagnostic reasoning process. Given initially only limited data, we leverage a (replaceable) NLP model to both support experts in their further data annotation with automatic suggestions, and we provide automatic feedback for students. We argue that because the central model consistently improves, our interactive approach encourages both students and instructors to recurrently use the tool, and thus accelerate the speed of data creation and annotation. We show results from two user studies on diagnostic reasoning in medicine and teacher education and outline how our system can be extended to further use cases.

Claudia Schulz, Christian M. Meyer, Jan Kiesewetter et al.

Many complex discourse-level tasks can aid domain experts in their work but require costly expert annotations for data creation. To speed up and ease annotations, we investigate the viability of automatically generated annotation suggestions for such tasks. As an example, we choose a task that is particularly hard for both humans and machines: the segmentation and classification of epistemic activities in diagnostic reasoning texts. We create and publish a new dataset covering two domains and carefully analyse the suggested annotations. We find that suggestions have positive effects on annotation speed and performance, while not introducing noteworthy biases. Envisioning suggestion models that improve with newly annotated texts, we contrast methods for continuous model adjustment and suggest the most effective setup for suggestions in future expert tasks.