most citedRefining a -nearest neighbor graph for a computationally efficient spectral clustering

44 citations · 56 across the 4 of their papers we have counts for

collaborators

6 papers

cs.LG20231 cited

Random projection tree similarity metric for SpectralNet

Mashaan Alshammari, John Stavrakakis, Adel F. Ahmed +1

SpectralNet is a graph clustering method that uses neural network to find an embedding that separates the data. So far it was only used with -nn graphs, which are usually constr…

cs.LG20238 cited

A parameter-free graph reduction for spectral clustering and SpectralNet

Mashaan Alshammari, John Stavrakakis, Masahiro Takatsuka

Graph-based clustering methods like spectral clustering and SpectralNet are very efficient in detecting clusters of non-convex shapes. Unlike the popular -means, graph-based clu…

cs.LG20233 cited

The Effect of Points Dispersion on the -nn Search in Random Projection Forests

Mashaan Alshammari, John Stavrakakis, Adel F. Ahmed +1

Partitioning trees are efficient data structures for -nearest neighbor search. Machine learning libraries commonly use a special type of partitioning trees called d-trees to…

cs.LG202344 cited

Refining a -nearest neighbor graph for a computationally efficient spectral clustering

Mashaan Alshammari, John Stavrakakis, Masahiro Takatsuka

Spectral clustering became a popular choice for data clustering for its ability of uncovering clusters of different shapes. However, it is not always preferable over other clusteri…

cs.LG2023

Random Projection Forest Initialization for Graph Convolutional Networks

Mashaan Alshammari, John Stavrakakis, Adel F. Ahmed +1

Graph convolutional networks (GCNs) were a great step towards extending deep learning to unstructured data such as graphs. But GCNs still need a constructed graph to work with. To…

cs.LG2023

Graph Construction using Principal Axis Trees for Simple Graph Convolution

Mashaan Alshammari, John Stavrakakis, Adel F. Ahmed +1

Graph Neural Networks (GNNs) are increasingly becoming the favorite method for graph learning. They exploit the semi-supervised nature of deep learning, and they bypass computation…