RankAxis: Towards a Systematic Combination of Projection and Ranking in Multi-Attribute Data Exploration

Qiangqiang Liu, Yukun Ren, Zhihua Zhu, Dai Li, Xiaojuan Ma, Quan Li

View presentation:2022-10-20T14:00:00ZGMT-0600Change your timezone on the schedule page
2022-10-20T14:00:00Z
Exemplar figure, described by caption below
(A) The data loader facilitates data selection; (B) The interactive projection view shows the projection distribution and guides analysts to explore the projection layout and directional semantics; (C1 – C5) The ranking tabular view summarizes the attribute contributions to the ranking and supports the deduction of the attribute weights based on user interaction, as well as compares different ranking schemes; (D) The comparative projection view analyzes the distribution of observations generated by different ranking schemes; (E) The ranking projection axis view compares the results of projection and ranking in the same context.

Prerecorded Talk

The live footage of the talk, including the Q&A, can be viewed on the session page, Interactive Dimensionality (High Dimensional Data).

Fast forward
Abstract

Projection and ranking are frequently used analysis techniques in multi-attribute data exploration. Both families of techniques help analysts with tasks such as identifying similarities between observations and determining ordered subgroups, and have shown good performances in multi-attribute data exploration. However, they often exhibit problems such as distorted projection layouts, obscure semantic interpretations, and non-intuitive effects produced by selecting a subset of (weighted) attributes. Moreover, few studies have attempted to combine projection and ranking into the same exploration space to complement each other's strengths and weaknesses. For this reason, we propose RankAxis, a visual analytics system that systematically combines projection and ranking to facilitate the mutual interpretation of these two techniques and jointly support multi-attribute data exploration. A real-world case study, expert feedback, and a user study demonstrate the efficacy of RankAxis.