The paper discusses the development and application of Kyurem, an interactive widget library designed to aid in human-centered knowledge acquisition and integration tasks within knowledge graphs (KGs). The tool provides a seamless environment within computational notebooks for data scientists to explore a KG, identify opportunities for new knowledge, and verify AI recommendations for integrating the acquired knowledge. Case studies conducted in real-world settings demonstrated Kyurem’s effectiveness in improving user experience and eliminating inefficiencies in knowledge acquisition and integration tasks.
Publication date: 7 Feb 2024
Project Page: https://doi.org/XXXXXXX.XXXXXXX
Paper: https://arxiv.org/pdf/2402.03291
