Jakob Smedegaard Andersen

• Master of Science
• PhD Student at HAW Hamburg

Text classification by large deep learning networks achieves high accuracy, but uses a lot of computing power and requires high resource investment. In the age of climate change, sustainable solutions are sought that can attain acceptable accuracy with less resource investment. In this paper, we investigate lightweight text classifiers and combine...

The development and deployment of systems using supervised machine learning (ML) remain challenging: mainly due to the limited reliability of prediction models and the lack of knowledge on how to effectively integrate human intelligence into automated decision-making. Humans involvement in the ML process is a promising and powerful paradigm to over...

To maximize the accuracy and increase the overall acceptance of text classifiers, we propose a framework for the efficient, in-operation moderation of classifiers' output. Our framework focuses on use cases in which F1-scores of modern Neural Networks classifiers (ca.~90%) are still inapplicable in practice. We suggest a semi-automated approach tha...

Visual perception is one of the most essential abilities for humans. This ability allows us to discover the world around us and to understand interdependencies with regard to both global context and particular concrete problem statement. We present in this paper an exploration of visualization and visual analytics approaches. Thereby, we focus on t...