Accelerating the k-means++ Algorithm by Using Geometric Information
This work provides an incremental improvement for users needing faster clustering in data analysis, particularly effective in low-dimensional and high-dimensional scenarios.
The paper tackled the problem of accelerating the k-means++ algorithm by incorporating geometric information like the Triangle Inequality and norm filters, resulting in reduced visited points and distance calculations, with speedup increasing with more clusters.
In this paper, we propose an acceleration of the exact k-means++ algorithm using geometric information, specifically the Triangle Inequality and additional norm filters, along with a two-step sampling procedure. Our experiments demonstrate that the accelerated version outperforms the standard k-means++ version in terms of the number of visited points and distance calculations, achieving greater speedup as the number of clusters increases. The version utilizing the Triangle Inequality is particularly effective for low-dimensional data, while the additional norm-based filter enhances performance in high-dimensional instances with greater norm variance among points. Additional experiments show the behavior of our algorithms when executed concurrently across multiple jobs and examine how memory performance impacts practical speedup.