TY - GEN
T1 - BICP
T2 - 25th ACM International Conference on Information and Knowledge Management, CIKM 2016
AU - Huang, Shengyu
AU - Candan, Kasim
AU - Sapino, Maria Luisa
N1 - Funding Information:
This work was supported by NSF grants 1339835, 1318788, and 1610282.
Publisher Copyright:
© 2016 Copyright held by the owner/author(s).
PY - 2016/10/24
Y1 - 2016/10/24
N2 - With many applications relying on multi-dimensional datasets for decision making, tensors (or multi-dimensional arrays) are emerging as a popular data representation to support diverse types of data, such as sensor streams and social networks. Consequently, tensor decomposition forms the basis for many data analysis and knowledge discovery tasks, from clustering, trend detection, anomaly detection, to correlation analysis. In applications where data evolves over time and the tensor-based analysis results need to be continuously maintained, re-computation of the whole tensor decomposition with each update will cause high computational costs and incur large memory overheads. In this paper, we propose a two-phase block-incremental CP-based tensor decomposition technique, BICP, that efficiently and effectively maintains tensor decomposition results in the presence of dynamically evolving tensor data. In its first phase, instead of repeatedly conducting ALS on each subtensor, BICP only revises the decompositions of the tensors that contain updated data. Moreover, when updates are relatively small with respect to the block size, BICP relies on a incremental factor tracking to avoid re-decomposition the updated sub-tensor. In its second phase, BICP limits the block-centric refinement process to only those blocks that are critical given the update. Experiment results show that the proposed method significantly reduces the execution time while assuring high accuracy.
AB - With many applications relying on multi-dimensional datasets for decision making, tensors (or multi-dimensional arrays) are emerging as a popular data representation to support diverse types of data, such as sensor streams and social networks. Consequently, tensor decomposition forms the basis for many data analysis and knowledge discovery tasks, from clustering, trend detection, anomaly detection, to correlation analysis. In applications where data evolves over time and the tensor-based analysis results need to be continuously maintained, re-computation of the whole tensor decomposition with each update will cause high computational costs and incur large memory overheads. In this paper, we propose a two-phase block-incremental CP-based tensor decomposition technique, BICP, that efficiently and effectively maintains tensor decomposition results in the presence of dynamically evolving tensor data. In its first phase, instead of repeatedly conducting ALS on each subtensor, BICP only revises the decompositions of the tensors that contain updated data. Moreover, when updates are relatively small with respect to the block size, BICP relies on a incremental factor tracking to avoid re-decomposition the updated sub-tensor. In its second phase, BICP limits the block-centric refinement process to only those blocks that are critical given the update. Experiment results show that the proposed method significantly reduces the execution time while assuring high accuracy.
UR - http://www.scopus.com/inward/record.url?scp=84996538084&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84996538084&partnerID=8YFLogxK
U2 - 10.1145/2983323.2983717
DO - 10.1145/2983323.2983717
M3 - Conference contribution
AN - SCOPUS:84996538084
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 1221
EP - 1230
BT - CIKM 2016 - Proceedings of the 2016 ACM Conference on Information and Knowledge Management
PB - Association for Computing Machinery
Y2 - 24 October 2016 through 28 October 2016
ER -