In this paper, we propose a method of fast comprehensive mining of coevolving high-order time series (FACets). It formulates high-order time series as tensors ...
The key idea is to use tensor factorization to address multi-aspect challenges, and perform careful regularizations to attack both contextual and temporal ...
In this paper, we propose a method of fast comprehensive mining of coevolving high-order time series (FACets). It formulates high-order time series as tensors ...
This paper proposes a comprehensive method, FACETS, to simultaneously model all three prominent challenges of mining time series data, using tensor ...
Our experimental evaluations on three real datasets demonstrate that our method (1) outperforms its competitors in two common data mining tasks (imputation and ...
Oct 14, 2021 · Bibliographic details on Facets: Fast Comprehensive Mining of Coevolving High-order Time Series.
Dive into the research topics of 'Facets: Fast comprehensive mining of coevolving high-order time series'. Together they form a unique fingerprint. Sort by ...
Facets: Fast Comprehensive Mining of Coevolving High-order Time Series KDD, 2015. KDD 2015 · DBLP · Scholar · DOI. Full names. Links ISxN. @inproceedings{KDD- ...
Facets : Fast Comprehensive Mining of Coevolving High-order Time Series. Cai, Yongjie;Tong, Hanghang;Fan, Wei;Ji, Ping;He, ...More. the 21th ACM SIGKDD ...
This paper proposes DCMF, a dynamic contextual matrix factorization algorithm, which outperforms its competitors, especially when there are lots of missing ...