Neighborhood hit for five synthetic data sets

Comparison of neighborhood hit (Qh) for DR and HD-SDR of the five different types of synthetic data used for Figure 3. Note that St-SNE, t-SNE, and SLMDS yield high Qh values near one for (a)–(c) and (e), which suggests that the corresponding labels of the k-size neighborhoods are well-preserved for HD-SDR. However, HD-SDR for sub-clustered data produces lower Qh compared with DR, as shown in (d), and this can be seen visually in Figure 3(a)–(d). More results including Qt, Qc, and Qj for the five synthetic data sets can be found in the supplemental materials.