# American Institute of Mathematical Sciences

September  2019, 1(3): 307-327. doi: 10.3934/fods.2019014

## Power weighted shortest paths for clustering Euclidean data

 1 Department of Mathematics, University of California, Los Angeles, Los Angeles CA 90095, USA 2 Department of Mathematics, University of Michigan, Ann Arbor MI 48109, USA

* Corresponding author: mckenzie@math.ucla.edu

Published  September 2019

Fund Project: The first author gratefully acknowledges the support of the Department of Mathematics, The University of Georgia, where the first author was a graduate student while this work was completed. The second author thanks the Department of Mathematics, The University of Michigan for their support. Both authors thank the anonymous reviewer for many useful suggestions.

We study the use of power weighted shortest path metrics for clustering high dimensional Euclidean data, under the assumption that the data is drawn from a collection of disjoint low dimensional manifolds. We argue, theoretically and experimentally, that this leads to higher clustering accuracy. We also present a fast algorithm for computing these distances.

Citation: Daniel Mckenzie, Steven Damelin. Power weighted shortest paths for clustering Euclidean data. Foundations of Data Science, 2019, 1 (3) : 307-327. doi: 10.3934/fods.2019014
##### References:

show all references

##### References:
Three sample geodesics in the power weighted shortest path metric with $p=2$, for the data set "Three Lines" (see §6). Observe how the geodesics consist of many small hops, instead of several large hops. The total lengths of the red and green paths are significantly smaller than the length of the blue path
All three synthetic data sets, projected into $\mathbb{R}^{2}$. From left to right: Three Lines, Three Moons and Three Circles
Varying $p$ and recording the accuracy of spectral clustering on the Three Lines data set, for three different values of the ambient dimension
Classification accuracy of spectral clustering. Note that $A^{(1)}$ represents using the Euclidean metric
 $A^{(f, 1)}$ $A^{(1)}$ $A^{(2)}$ $A^{(10)}$ $A^{(\infty)}$ 3 Lines $66.11\pm 0.94\%$ $66.35 \pm 3.73\%$ $66.87 \pm 3.37\%$ $95.38\pm 9.22\%$ $\bf{95.38 \pm 9.1\%}$ 3 Moons $85.90 \pm 1.13\%$ $94.40 \pm 1.48\%$ $94.40 \pm 1.48\%$ $\bf{96.20 \pm 1.76\%}$ $94.35 \pm 3.34\%$ 3 Circles $51.87 \pm 0.00\%$ $51.93 \pm 0.32\%$ $51.94 \pm 0.36\%$ $71.22 \pm 9.50\%$ $\bf{73.61 \pm 10.47\%}$ $\mathtt{ DrivFace}$ $78.88\%$ $71.62\%$ $71.62\%$ $74.71\%$ $\bf{85.38\%}$ $\mathtt{ COIL-20}$ $63.24\%$ $75.28\%$ $\bf{78.61\%}$ $77.45\%$ $60.92\%$ $\mathtt{ OptDigits}$ $77.73\%$ $91.49\%$ $\bf{91.54\%}$ $88.39\%$ $83.17\%$ $\mathtt{ USPS}$ $48.65\%$ $65.05\%$ $65.02\%$ $76.20\%$ $\bf{77.92\%}$ $\mathtt{ MNIST}$ - $76.11\%$ $75.63\%$ $84.54\%$ $\bf{86.77\%}$
 $A^{(f, 1)}$ $A^{(1)}$ $A^{(2)}$ $A^{(10)}$ $A^{(\infty)}$ 3 Lines $66.11\pm 0.94\%$ $66.35 \pm 3.73\%$ $66.87 \pm 3.37\%$ $95.38\pm 9.22\%$ $\bf{95.38 \pm 9.1\%}$ 3 Moons $85.90 \pm 1.13\%$ $94.40 \pm 1.48\%$ $94.40 \pm 1.48\%$ $\bf{96.20 \pm 1.76\%}$ $94.35 \pm 3.34\%$ 3 Circles $51.87 \pm 0.00\%$ $51.93 \pm 0.32\%$ $51.94 \pm 0.36\%$ $71.22 \pm 9.50\%$ $\bf{73.61 \pm 10.47\%}$ $\mathtt{ DrivFace}$ $78.88\%$ $71.62\%$ $71.62\%$ $74.71\%$ $\bf{85.38\%}$ $\mathtt{ COIL-20}$ $63.24\%$ $75.28\%$ $\bf{78.61\%}$ $77.45\%$ $60.92\%$ $\mathtt{ OptDigits}$ $77.73\%$ $91.49\%$ $\bf{91.54\%}$ $88.39\%$ $83.17\%$ $\mathtt{ USPS}$ $48.65\%$ $65.05\%$ $65.02\%$ $76.20\%$ $\bf{77.92\%}$ $\mathtt{ MNIST}$ - $76.11\%$ $75.63\%$ $84.54\%$ $\bf{86.77\%}$
Run time of spectral clustering, in seconds. Note that this includes the time rquired to construct the similarity matrix. $A^{(1)}$ represents using the Euclidean metric
 $A^{(f, 1)}$ $A^{(1)}$ $A^{(2)}$ $A^{(10)}$ $A^{(\infty)}$ 3 Lines $0.32$ $0.16$ $1.20$ $1.22$ $1.22$ 3 Moons $0.33$ $0.17$ $1.31$ $1.30$ $1.36$ 3 Circles $0.35$ $0.16$ $1.00$ $1.06$ $1.07$ $\mathtt{ DrivFace}$ $0.37$ $1.24$ $1.55$ $1.64$ $1.64$ $\mathtt{ COIL-20}$ $0.57$ $0.72$ $1.57$ $1.82$ $1.78$ $\mathtt{ OptDigits}$ $5.40$ $1.41$ $5.28$ $5.58$ $5.67$ $\mathtt{ USPS}$ $27.40$ $17.12$ $26.75$ $22.78$ $23.79$ $\mathtt{ MNIST}$ - $2060.23$ $2031.38$ $1554.15$ $1613.41$
 $A^{(f, 1)}$ $A^{(1)}$ $A^{(2)}$ $A^{(10)}$ $A^{(\infty)}$ 3 Lines $0.32$ $0.16$ $1.20$ $1.22$ $1.22$ 3 Moons $0.33$ $0.17$ $1.31$ $1.30$ $1.36$ 3 Circles $0.35$ $0.16$ $1.00$ $1.06$ $1.07$ $\mathtt{ DrivFace}$ $0.37$ $1.24$ $1.55$ $1.64$ $1.64$ $\mathtt{ COIL-20}$ $0.57$ $0.72$ $1.57$ $1.82$ $1.78$ $\mathtt{ OptDigits}$ $5.40$ $1.41$ $5.28$ $5.58$ $5.67$ $\mathtt{ USPS}$ $27.40$ $17.12$ $26.75$ $22.78$ $23.79$ $\mathtt{ MNIST}$ - $2060.23$ $2031.38$ $1554.15$ $1613.41$
 [1] Liping Tang, Ying Gao. Some properties of nonconvex oriented distance function and applications to vector optimization problems. Journal of Industrial & Management Optimization, 2021, 17 (1) : 485-500. doi: 10.3934/jimo.2020117 [2] Cheng Peng, Zhaohui Tang, Weihua Gui, Qing Chen, Jing He. A bidirectional weighted boundary distance algorithm for time series similarity computation based on optimized sliding window size. Journal of Industrial & Management Optimization, 2021, 17 (1) : 205-220. doi: 10.3934/jimo.2019107

Impact Factor:

## Tools

Article outline

Figures and Tables