PMID- 33090026
OWN - NLM
STAT- MEDLINE
DCOM- 20210930
LR  - 20210930
IS  - 2167-647X (Electronic)
IS  - 2167-6461 (Linking)
VI  - 8
IP  - 5
DP  - 2020 Oct
TI  - LTSpAUC: Learning Time-Series Shapelets for Partial AUC Maximization.
PG  - 391-411
LID - 10.1089/big.2020.0069 [doi]
AB  - Shapelets are discriminative segments used to classify time-series instances. 
      Shapelet methods that jointly learn both classifiers and shapelets have been 
      studied in recent years because such methods provide both interpretable results 
      and superior accuracy. The partial area under the receiver operating 
      characteristic curve (pAUC) for a low range of false-positive rates (FPR) is an 
      important performance measure for practical cases in industries such as medicine, 
      manufacturing, and maintenance. In this article, we propose a method that jointly 
      learns both shapelets and a classifier for pAUC optimization in any FPR range, 
      including the full AUC. In addition, we propose the following two extensions for 
      shapelet methods: (1) reducing algorithmic complexity in time-series length to 
      linear time and (2) explicitly determining the classes that shapelets tend to 
      match. Comparing with state-of-the-art learning-based shapelet methods, we 
      demonstrated the superiority of pAUC on UCR time-series data sets and its 
      effectiveness in industrial case studies from medicine, manufacturing, and 
      maintenance.
FAU - Yamaguchi, Akihiro
AU  - Yamaguchi A
AD  - System AI Laboratory, Corporate R&D Center, Toshiba Corporation, Kawasaki, Japan.
FAU - Maya, Shigeru
AU  - Maya S
AD  - System AI Laboratory, Corporate R&D Center, Toshiba Corporation, Kawasaki, Japan.
FAU - Maruchi, Kohei
AU  - Maruchi K
AD  - System AI Laboratory, Corporate R&D Center, Toshiba Corporation, Kawasaki, Japan.
FAU - Ueno, Ken
AU  - Ueno K
AD  - System AI Laboratory, Corporate R&D Center, Toshiba Corporation, Kawasaki, Japan.
LA  - eng
PT  - Journal Article
PL  - United States
TA  - Big Data
JT  - Big data
JID - 101631218
SB  - IM
MH  - Algorithms
MH  - *Area Under Curve
MH  - *False Positive Reactions
MH  - *Learning
MH  - *Software
OTO - NOTNLM
OT  - AUC
OT  - partial AUC
OT  - shapelets
OT  - time series
OT  - time-series classification
EDAT- 2020/10/23 06:00
MHDA- 2021/10/01 06:00
CRDT- 2020/10/22 12:15
PHST- 2020/10/22 12:15 [entrez]
PHST- 2020/10/23 06:00 [pubmed]
PHST- 2021/10/01 06:00 [medline]
AID - 10.1089/big.2020.0069 [doi]
PST - ppublish
SO  - Big Data. 2020 Oct;8(5):391-411. doi: 10.1089/big.2020.0069.