期刊
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
卷 28, 期 1, 页码 540-550出版社
IEEE COMPUTER SOC
DOI: 10.1109/TVCG.2021.3114870
关键词
Data visualization; Visualization; Task analysis; Data analysis; Topology; Image color analysis; Dimensionality reduction; Dimensionality reduction; embedding; augmented projections; point set contours; explainable artificial intelligence
资金
- Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) [252408385 -IRTG 2057]
This paper discusses the importance of embeddings of high-dimensional data and the difficulty in explaining them. By introducing Non-Linear Embeddings Surveyor (NoLiES) and a new augmentation strategy called rangesets, users are able to quickly observe the structure and detect outliers.
Embeddings of high-dimensional data are widely used to explore data, to verify analysis results, and to communicate information. Their explanation, in particular with respect to the input attributes, is often difficult. With linear projects like PCA the axes can still be annotated meaningfully. With non-linear projections this is no longer possible and alternative strategies such as attribute-based color coding are required. In this paper, we review existing augmentation techniques and discuss their limitations. We present the Non-Linear Embeddings Surveyor (NoLiES) that combines a novel augmentation strategy for projected data (rangesets) with interactive analysis in a small multiples setting. Rangesets use a set-based visualization approach for binned attribute values that enable the user to quickly observe structure and detect outliers. We detail the link between algebraic topology and rangesets and demonstrate the utility of NoLiES in case studies with various challenges (complex attribute value distribution, many attributes, many data points) and a real-world application to understand latent features of matrix completion in thermodynamics.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据