Visual exploration analysis method and system for high-dimensional data subspace

The invention discloses a visual exploration analysis method and system for high-dimensional data subspace, and the method comprises the steps: S100, searching valuable subspaces from a high-dimensional data set, and extracting the features of the valuable subspaces, the features comprising subspace similarity, subspace clustering, a data mode, a dimension mode and a representative subspace; S200, constructing a subspace map based on the extracted features, including determining anchor point positions and a subspace traversal sequence, determining layout in a grid tiling manner, and adding an additional map metaphor; and S300, drawing a subspace map and other auxiliary views, and completing subspace exploration through interaction with the user. The user is supported to efficiently complete exploration analysis of the high-dimensional data subspace in a visual form of the map metaphor. 本发明公开了一种高维数据子空间可视化探索分析方法和系统，方法包括：S100、从高维数据集中搜索有价值的子空间并提取其特征，特征包括子空间相似度、子空间聚类、数据模式、维度模式和代表性子空间；S200、基于提取的特征，构建子空间地图，包括确定锚点位置和