Portal de Periódicos da CAPES

Interactive Visual Exploration of Big Relational Datasets

2022; Taylor & Francis; Volume: 39; Issue: 10 Linguagem: Inglês

10.1080/10447318.2022.2073007

1532-7590

Katerina Vitsaxaki, Stavroula Ntoa, George Margetis, Nicolas Spyratos,

Scientific Computing and Data Management

We propose and implement a tool for the access, analysis, and visual exploration of very large relational databases and data warehouses. To this end, we represent the database as a graph in which each table is represented by a tree rooted at the key and in which there is an edge from table T to table T' if there is an attribute of T which is a foreign key of T'. The interface allows the user to define analytic queries by clicking on nodes and edges of the graph. The user actions are then translated into SQL group-by queries that are evaluated by the underlying database management system. Additionally, the interface allows the user to visualize the analysis results, and to interact with visualizations in order to explore the database content.