KG-PRE-view: Democratizing a TVCG Knowledge Graph through Visual Explorations

IEEE Transactions on Visualization and Computer Graphics (TVCG) publishes cutting-edge research in the fields of visualization, computer graphics, and virtual and augmented realities. Within the TVCG ecosystem, different stakeholders make decisions based on available information related to TVCG almost on a daily basis. The decisions involve various tasks such as the retrieval of research ideas and trends, the invitation of peer reviewers, and the selection of editorial board members, just to name a few. To make well-informed decisions in these contexts, a data-driven approach is necessary. However, the current IEEE digital library only provides access to individual papers. Transforming this wealth of data into valuable insights is a daunting task, requiring specialized expertise and effort in tasks such as data crawling, cleaning, analysis, and visualizations. To address the needs of the community in facilitating more efficient and transparent decision-making, we construct and publicly release a TVCG knowledge graph (TVCG-KG). TVCG-KG is a structured representation of heterogeneous information, including the metadata of each publication such as author, affiliation, title, and semantic information such as method, task, data. Despite the widespread use of KGs in various downstream applications, a noticeable gap exists in the visualization literature regarding the full exploitation of the rich semantics embedded within KGs. While it might seem intuitive to just employ interactive graph-based visualization for KGs, we propose that knowledge discovery over KG is a series of visual exploratory tasks that can benefit from using multiple visualization techniques and designs. We conducted an evaluation of TVCG-KG quality and demonstrated its practical utility through several real-world cases. Our data and code are accessible via the following URL: https://github.com/yasmineTYM/TVCG-KG.git.