TY - GEN
T1 - T2-Net
T2 - 20th IEEE International Conference on Data Mining, ICDM 2020
AU - Zhang, Denghui
AU - Liu, Yanchi
AU - Cheng, Wei
AU - Zong, Bo
AU - Ni, Jingchao
AU - Chen, Zhengzhang
AU - Chen, Haifeng
AU - Xiong, Hui
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11
Y1 - 2020/11
N2 - Accurate air turbulence forecasting can help airlines avoid hazardous turbulence, guide the routes that keep passengers safe, maximize efficiency, and reduce costs. Traditional turbulence forecasting approaches heavily rely on painstakingly customized turbulence indexes, which are less effective in dynamic and complex weather conditions. The recent availability of high-resolution weather data and turbulence records allows more accurate forecasting of the turbulence in a data-driven way. However, it is a non-trivial task for developing a machine learning based turbulence forecasting system due to two challenges: (1) Complex spatio-temporal correlations, turbulence is caused by air movement with complex spatio-temporal patterns, (2) Label scarcity, very limited turbulence labels can be obtained. To this end, in this paper, we develop a unified semi-supervised framework, T2-Net, to address the above challenges. Specifically, we first build an encoder-decoder paradigm based on the convolutional LSTM to model the spatio-temporal correlations. Then, to tackle the label scarcity problem, we propose a novel Dual Label Guessing method to take advantage of massive unlabeled turbulence data. It integrates complementary signals from the main Turbulence Forecasting task and the auxiliary Turbulence Detection task to generate pseudo-labels, which are dynamically utilized as additional training data. Finally, extensive experimental results on a real-world turbulence dataset validate the superiority of our method on turbulence forecasting.
AB - Accurate air turbulence forecasting can help airlines avoid hazardous turbulence, guide the routes that keep passengers safe, maximize efficiency, and reduce costs. Traditional turbulence forecasting approaches heavily rely on painstakingly customized turbulence indexes, which are less effective in dynamic and complex weather conditions. The recent availability of high-resolution weather data and turbulence records allows more accurate forecasting of the turbulence in a data-driven way. However, it is a non-trivial task for developing a machine learning based turbulence forecasting system due to two challenges: (1) Complex spatio-temporal correlations, turbulence is caused by air movement with complex spatio-temporal patterns, (2) Label scarcity, very limited turbulence labels can be obtained. To this end, in this paper, we develop a unified semi-supervised framework, T2-Net, to address the above challenges. Specifically, we first build an encoder-decoder paradigm based on the convolutional LSTM to model the spatio-temporal correlations. Then, to tackle the label scarcity problem, we propose a novel Dual Label Guessing method to take advantage of massive unlabeled turbulence data. It integrates complementary signals from the main Turbulence Forecasting task and the auxiliary Turbulence Detection task to generate pseudo-labels, which are dynamically utilized as additional training data. Finally, extensive experimental results on a real-world turbulence dataset validate the superiority of our method on turbulence forecasting.
KW - Semi-supervised learning
KW - Spatio-temporal modeling
KW - Turbulence forecasting
UR - http://www.scopus.com/inward/record.url?scp=85100875806&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100875806&partnerID=8YFLogxK
U2 - 10.1109/ICDM50108.2020.00182
DO - 10.1109/ICDM50108.2020.00182
M3 - Conference contribution
AN - SCOPUS:85100875806
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 1388
EP - 1393
BT - Proceedings - 20th IEEE International Conference on Data Mining, ICDM 2020
A2 - Plant, Claudia
A2 - Wang, Haixun
A2 - Cuzzocrea, Alfredo
A2 - Zaniolo, Carlo
A2 - Wu, Xindong
Y2 - 17 November 2020 through 20 November 2020
ER -