SemanticAxis: exploring multi-attribute data by semantic construction and ranking analysis

Abstract

Ranking rows D top institutions in the Vis area top institutions in the Graphics area institutions that perform similar in these two areas institutions that have no papers in either area Graphics Utah Stony Brook UC-Davis Stanford MIT attribute row filtering results row weighted ranking row Ohio State Stuttgart Stanford filter brush Visualization row Graphics row Utah

Incorporation of Human Knowledge Into Data Embeddings to Improve Pattern Significance and Interpretability

“…Xia et al [80] proposed descriptors to construct the subspace to reveal low-dimensional structures unobservable in the original dimensional space. Many approaches allow the user to construct dimensions of the subspace interactively [25,32,39]. Each subspace dimension is the linear combination of many original dimensions.…”

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

“…On the other hand, OLI enables analysts to manipulate the observed data directly, insulating them from the complexity of the underlying mathematical model. For example, Li et al [25] proposed SemanticAxis that enables analysts to reconstruct projections by directly modifying attribute weights, which clearly falls under a PI. They also supported the creation of a semantic axis by selecting two sets of data observations.…”