An interactive visualization interface for studying egocentric, categorical, contact diary datasets
Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining
| Hi-index | 0.00 |
Scatter plots are one of the most powerful techniques for visualizing relationships between two continuous variables. Using scatter plots, it is easy to find how one variable is affected by another. However, scatter plots may have a high degree of overlap, and therefore, important local patterns and trends may be hidden. Generalized scatter plots provide overlap-distortion optimized views, but the point repositioning algorithm used for avoiding overlap does not take the local structure of the data into account. In this paper, we enhance scatter plots using ellipsoid point placement and a cluster shading algorithm. In particular, we use local correlations to compute the rotation and aspect ratios of the ellipsoids used for the point placement, and add shading to the point clusters to visually encode the points' original locations. The effect of the shading and lighting can be controlled by the user.