This repository provides the Matlab source code for two large-scale clustering algorithms, namely, Ultra-Scalable Spectral Clustering (U-SPEC) and Ultra-Scalable Ensemble Clustering (U-SENC), both of which have nearly linear time and space complexity and are capable of robustly and efficiently partitioning ten-million-level nonlinearly-separable datasets on a PC with 64GB memory.

If you find this repository helpful for your research, please cite the paper below.

Dong Huang, Chang-Dong Wang, Jian-Sheng Wu, Jianhuang Lai, and Chee-Keong Kwoh.
Ultra-Scalable Spectral Clustering and Ensemble Clustering, 
IEEE Transactions on Knowledge and Data Engineering (TKDE), 2020, vol.32, no.6, pp.1212-1226. 
DOI: https://doi.org/10.1109/TKDE.2019.2903410

In this repository, we provide the files of the five real-world datasets, namely, PenDigits, USPS, Letters, MNIST, and Covertype. We also provide the MATLAB code to reproduce the five large-scale synthetic datasets used in our paper.

• To generate the five large-scale synthetic datasets, you can call the synthesizeLargescaleDatasets function, which has just one input parameter. Note that this input parameter can be set to one of the five data names:

  • 'TB1M'
  • 'SF2M'
  • 'CC5M'
  • 'CG10M'
  • 'Flower20M'

  Example (to synthesize the CC5M dataset):

   synthesizeLargescaleDatasets('CC5M');
   % The synthesized dataset will be saved in 'data_CC5M.mat'.
  

• To generate the five synthetic datasets with arbitrary sizes, you can call the synthesizeLargescaleDatasets_withArbitrarySizes function, which has two input parameters, that is

  • dataName: can be one of the five names: 'TB', 'SF', 'CC', 'CG', 'Flower'
  • dataSize: can be set to any integers, provided that you have enough space to save the data.

  Example (to synthesize a CG dataset with one million points):

   dataName = 'CG';
   dataSize = 1000000;
   synthesizeLargescaleDatasets_withArbitrarySizes(dataName, dataSize); 
   % The synthesized dataset will be saved in 'data_CG_1000000.mat'.
  

Don't hesitate to contact me if you have any questions regarding this work.
Email: huangdonghere at gmail dot com
Website: https://www.researchgate.net/publication/330760669