Document Type
Thesis
Date of Degree Completion
Fall 2024
Degree Name
Master of Science (MS)
Department
Computational Science
Committee Chair
Dr. Boris Kovalerchuk
Second Committee Member
Dr. Razvan Andonie
Third Committee Member
Dr. Szilard VAJDA
Abstract
This research advances interpretable machine learning (ML) by introducing hyperblocks (HBs) as a structured, rule-based approach for creating transparent and accurate models using meaningful numeric attributes directly interpretable to end users. Key techniques, including Parallel Hyperblock Creation, Interactive Hyperblock Creation, Level n Hyperblock Creation, and k-Nearest Neighbor Hyperblock, provide a framework that ensures domain experts can meaningfully engage with the model’s decision-making process through lossless visualizations using General Line Coordinates (GLC). Case studies with the Wisconsin Breast Cancer and MNIST datasets demonstrated HBs' effectiveness in handling high-risk and complex classification tasks, offering interpretable accuracy that traditional models struggle to achieve. In the Wisconsin Breast Cancer study, HBs achieved an accuracy comparable to standard ML methods while providing an interpretable framework for cancer diagnosis, where model trust is critical. In MNIST, HBs showed their ability to scale to larger datasets while maintaining a high level of interpretability. Finally, the Visual Knowledge Discovery (VKD) process, central to this approach, allows experts to adjust model parameters in real time, promoting human-centered insights and collaboration. Overall, this work presents HBs and VKD as powerful tools for interpretable, high-stakes ML applications, supporting transparency, accuracy, and domain relevance.
Recommended Citation
Huber, Lincoln, "Human-Centered Machine Learning with Interpretable Visual Knowledge Discovery" (2024). All Master's Theses. 1991.
https://digitalcommons.cwu.edu/etd/1991
Included in
Artificial Intelligence and Robotics Commons, Graphics and Human Computer Interfaces Commons