Visual analytics of set data for knowledge discovery and member selection support

Visual analytics (VA) is a visually assisted exploratory analysis approach in which knowledge discovery executed interactively between the user and system human-centered manner. The purpose of this study to develop method for VA set data aimed at supporting member selection. A typical target application visual support team selection, by users can analyze past teams examine candidate lineups new teams. Because there are several difficulties, such as combinatorial explosion problem, developing challenging. In study, we first define requirements that should satisfy clarify accompanying challenges. Then propose data, satisfies requirements. key idea model generation process sets their outputs using manifold network model. proposed visualizes relevant factors topographic maps on various information visualized. Furthermore, bidirectional interface, indicate targets interest these maps. We demonstrate applying it basketball teams, compare with benchmark outcome prediction lineup reconstruction tasks. be adapted individual cases extending structure, general practical systems built.