Skip to Main content Skip to Navigation
Conference papers

ARViz: Interactive Visualization of Association Rules for RDF Data Exploration

Abstract : Association rule mining often leads the analyst into a rough rummaging process to identify rules that are relevant to understand specific problems. We propose a visualization interface to assist the rule selection process and evaluate it on an RDF knowledge graph derived from the COVID-19 Open Research Dataset. The user interface supports data exploration with focus on the overview of rules through a scatter plot, subsets of rules through a chord diagram chart, and itemsets through an association graph which is dynamically created by entering an item of interest (i.e. a named entity). Further, the analyst can interactively recover a list of publications containing the named entities involved in a particular rule. Among the original aspects of our approach, we highlight the representation of attributes describing measures of interest (i.e. confidence and interestingness), a visual indication of existence (or not) of symmetry in association rules, the exploration of subsets of rules according to clusters of publications and named entities, and an interactive prompting that aims at expanding the discovery of named entities within selected association rules. We assess our approach through a semi-structured interview involving experts in the domains of data mining and biomedicine, whose feedback could assist the refinement of the visual and interaction tools.
Complete list of metadata
Contributor : Aline Menin Connect in order to contact the contributor
Submitted on : Tuesday, July 20, 2021 - 11:02:55 AM
Last modification on : Friday, August 5, 2022 - 3:50:30 AM
Long-term archiving on: : Thursday, October 21, 2021 - 6:34:11 PM


Files produced by the author(s)



Aline Menin, Lucie Cadorel, Andrea G. B. Tettamanzi, Alain Giboin, Fabien Gandon, et al.. ARViz: Interactive Visualization of Association Rules for RDF Data Exploration. IV 2021 - 25th International Conference Information Visualisation, Jul 2021, Melbourne / Virtual, Australia. pp.13-20, ⟨10.1109/IV53921.2021.00013⟩. ⟨hal-03292140⟩



Record views


Files downloads