1.3 Research Contributions
In this dissertation, we present a scalable physics-based data modeling framework
(see Figure 1.1) for unsupervised high-dimensional applications. Particularly
speaking, our contributions in this research include:
• Density-Aware Clustering based on Aggregated Heat Kernel and Its
Transformation
We proposed a diffusion-based Aggregated Heat Kernel (AHK) to improve
the clustering stability, and a Local Density Affinity Transformation (LDAT)
to correct the bias originating from different cluster densities. AHK aggregately
models the heat diffusion traces along all the time scales, so it ensures