Overview
DCube (
Diskbased
Denseblock
Detection) is an algorithm for detecting dense blocks in webscale tensors.
DCube has the following properties:
 Scalable: DCube handles large data not fitting in memory or even on a disk.
 Fast: Even when data fit in memory, DCube outperforms its competitors in terms of speed.
 Accurate: DCube gives high accuracy in realworld data as well as theoretical accuracy guarantees.
Papers
DCube is described in the following papers:

DCube: DenseBlock Detection in TerabyteScale Tensors
Kijung Shin, Bryan Hooi, Jisu Kim, and Christos Faloutsos.
The 10th ACM International Conference on Web Search and Data Mining (WSDM) 2017, Cambridge, UK
[PDF] [Supplementary Document] [BIBTEX]

OutofCore and Distributed Algorithms for Dense Subtensor Mining (Longer ver.)
Kijung Shin, Bryan Hooi, Jisu Kim, and Christos Faloutsos.
[Arxiv]
Code
The source code used in the papers is available.
[ver.1 (WSDM)]
[ver.2 (TKDE)]
Datasets
People