@article{oai:ipsj.ixsq.nii.ac.jp:02009109,
 author = {千原,直己 and 松原,靖子 and 藤原,廉 and 櫻井,保志 and Naoki Chihara and Yasuko Matsubara and Ren Fujiwara and Yasushi Sakurai},
 issue = {2},
 journal = {情報処理学会論文誌データベース（TOD）},
 month = {Apr},
 note = {本論文では，大規模時系列データストリーム中の時間変化する因果関係の抽出および将来予測を同時に行うための最新手法ModePlaitを提案する．提案手法は以下の優れた特性をすべて満たす．(a)時々刻々と変化する環境の移り変わりに従って変化する因果関係を明らかにする．(b)時間変化する因果関係の抽出および将来予測を同時かつ正確に行う．(c)計算時間は時系列データストリーム全体の長さに依存せず，高速に処理を行う．人工データおよび実データを用いた評価実験により，提案手法が最新の既存手法に比べて因果探索，将来予測の両方の観点において高精度であること，そして，計算効率の良い高速な処理が可能であることを明らかにした．, We study the novel problem of discovering the time-varying cause-and-effect relationships across transitions of dynamical patterns in multivariate co-evolving data streams. To solve such a problem, we present a streaming method, ModePlait, which is designed for modeling such causal relationships (i.e., time-evolving causality) and forecasting their future values. ModePlait has the following desirable properties: (a) Effective: it discovers the time-evolving causality in multivariate co-evolving data streams by detecting the transitions of distinct dynamical patterns adaptively. (b) Accurate: it enables both the discovery of time-evolving causality and the forecasting of future values in a streaming fashion. (c) Scalable: our algorithm does not depend on data stream length and thus is applicable to very large sequences. Extensive experiments on both synthetic and real-world datasets demonstrate that our proposed model outperforms state-of-the-art methods in terms of discovering the time-evolving causality as well as forecasting.},
 pages = {70--81},
 title = {時間変化する因果関係の抽出に基づいた高速将来予測},
 volume = {19},
 year = {2026}
}