Overview
How can we extract hidden relations from a tensor and a matrix data simultaneously in a fast, accurate, and scalable way? Coupled matrix-tensor factorization (CMTF) is an important tool for the purpose. Designing an accurate and efficient CMTF method has become more crucial as the size and dimension of real-world data are growing explosively. However, existing methods for CMTF suffer from lack of accuracy, slow running time, and limited scalability.
In this paper, we propose
S3CMTF, a fast, accurate, and scalable CMTF method. In contrast to previous methods which do not support sparse tensors or do not model complicated relationships between factors,
S3CMTF provides sparse Tucker factorization by carefully deriving gradient update rules. We also show that lock-free parallel SGD is useful for
S3CMTF in multi-core shared memory systems.
S3CMTF further boosts the performance by carefully storing intermediate computation and reusing them. We theoretically and empirically show that
S3CMTF is the fastest, outperforming existing methods. Experimental results show that
S3CMTF is up to 989x faster than existing methods while providing the best accuracy.
S3CMTF shows linear scalability on the number of data entries and the number of cores. In addition, we apply
S3CMTF to Yelp recommendation tensor data coupled with 3 additional matrices to discover interesting patterns.
Paper
-
S3CMTF: Fast, Accurate, and Scalable Method for Sparse Coupled Matrix-Tensor Factorization
Dongjin Choi, Jun-Gi Jang, and U Kang.
PLOS ONE 14(6): e0217316.
[PDF] [BIB]
Code
The source codes used in the paper are available.
Comparison
Comparison of our proposed S
3CMTF and existing CMTF methods. S
3CMTF outperforms the state of-the-art single machine CMTF methods in terms of time, accurracy, and speed.
Dataset
People