Kaitai Tong

Kaitai Tong

Auto-generated profile, based on publicly available information

Kaitai Tong

Institute for Resources, Environment and Sustainability, University of British Columbia, 2202 Main Mall, Vancouver, BC V6T 1Z4, Canada

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 January 2021 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Verified
Journal Article
Closed access
Teng LiTeng Li
Teng Li
Yuxin PanYuxin Pan
Yuxin Pan
KTKaitai Tong
Kaitai Tong
CVCarlos Estuardo Ventura
Carlos Estuardo Ventura
CSClarence de Silva
Clarence de Silva
In structural health monitoring (SHM), measuring and evaluating structural dynamic responses are critical for safety management of civil infrastructures. Particularly, online forecasting of the structural responses under extreme external loading conditions (e.g., earthquakes) takes a significant role in SHM to provide early warning and ensure safe operation. In practice, complex causality and intrinsic interactions between seismic excitation and structural response make it challenging to establish a reliable predictive scheme. The present paper proposes a novel deep recurrent neural network (RNN) model implemented in the architecture of a time-series attention-based RNN encoder–decoder (TSA-RNN-ED), for predictive analysis of structural responses under seismic excitation. In the proposed data-driven model, upcoming sequential responses are predicted through sequence-to-sequence learning from historical multivariate time-series signals. A time-series attention mechanism is proposed to exploit the heterogeneous, but directly related, hidden features between the seismic loads and the corresponding structural responses. The proposed architecture can reliably regress excitation-response interactions to predict dynamic responses subjected to future earthquakes while satisfying the need of real-time forecasting for on-site practical implementation. This article systematically evaluates the proposed model by using two real-world structural cases: 1) the tallest building in China, the Shanghai Tower and 2) a woodframe classroom on a shake table at the University of British Columbia in Vancouver, Canada. The experimental results demonstrate the accurate and efficient performance of the proposed methodology in forecasting the seismic responses of the structures under investigation.

22 July 2025 in Earth System Science Data
Verified
Journal Article
Open access
PDF file
ZMZia Mehrabi
Zia Mehrabi
KTKaitai Tong
Kaitai Tong
JFJulie Fortin
Julie Fortin
RSRadost Stanimirova
Radost Stanimirova
MFMark Friedl
Mark Friedl
NRNavin Ramankutty
Navin Ramankutty
While there are many global geospatial datasets representing the extent of agriculture, they predominantly represent croplands. Only a couple of global data products represent the full global agricultural footprint, including both cropland and pastures. Our own research team's most recent complete publicly available agricultural land cover dataset, including both croplands, and pastures, represents circa 2000. These data, distributed on a graticule of 5 arcmin (∼ 10 km2 at the Equator), have been integrated into a considerable number and diversity of research studies, modelling, data science, and media applications. Further, users of these data have been interested in them for studying a variety of issues such as land use, food security, climate change, and biodiversity loss. Here we present an updated dataset on the global distribution of agricultural lands (cropland and pasture) circa 2015 (15 years on since the initial study). Past studies that have constructed such datasets have been one-off exercises that have been infrequently repeated due to the amount of effort required. Therefore, in this work, we developed a transparent and reproducible approach to update our data product while also enabling easier reproduction of future datasets. We distribute our 2015 product at the same resolution and formats as the prior product, and accompany it with a full set of replicable code and data for reconstruction. In this article we explain how the data was constructed, with links to the permanent DOIs where the data can be readily downloaded by the user community (Mehrabi et al., 2024; https://doi.org/10.5281/zenodo.11540553).

18 September 2019 in IEEE Systems Journal
Verified
Journal Article
Closed access
Teng LiTeng Li
Teng Li
KTKaitai Tong
Kaitai Tong
Min XiaMin Xia
Min Xia
BLBing Li
Bing Li
CSClarence de Silva
Clarence de Silva
This article investigates the problem of information-based sampling design and path planning for a mobile sensing network to predict scalar fields of monitored environments. A hierarchical framework with a built-in Gaussian Markov random field model is proposed to provide adaptive sampling for efficient field reconstruction. In the proposed framework, a nonmyopic planner is operated at a sink to navigate the mobile sensing agents in the field to the sites that are most informative. Meanwhile, a myopic planner is carried out on board each agent. A tradeoff between computationally intensive global optimization and efficient local greedy search is incorporated into the system. The mobile sensing agents can be scheduled online through an anytime algorithm to visit and observe the high-information sites. Experiments on both synthetic and real-world datasets are used to demonstrate the feasibility and efficiency of the proposed planner in model exploitation and adaptive sampling for environmental field mapping.

14 August 2024 in EGUsphere
Verified
Preprint
Open access
PDF file
ZMZia Mehrabi
Zia Mehrabi
KTKaitai Tong
Kaitai Tong
JFJulie Fortin
Julie Fortin
RSRadost Stanimirova
Radost Stanimirova
MFMark Friedl
Mark Friedl
NRNavin Ramankutty
Navin Ramankutty
While there are many global geospatial datasets representing the extent of agriculture, they predominantly represent croplands. Only a couple of global data products represent the full global agricultural footprint, including pastures. Our own research team’s most recent complete publicly available agricultural land cover dataset, including both croplands and pastures, represent circa 2000. These data, distributed on a graticule of 5 arcminutes (~10 km2 at the equator), have been integrated into a considerable number and diversity of research studies, modeling, data science and media applications. Further, users of these data have been interested in them for studying a variety of issues such as land use, food security, climate change and biodiversity loss. Here we present an updated dataset on the global distribution of agricultural lands (cropland and pasture) circa 2015 (15 years on since the initial study). Past studies that have constructed such datasets have been one-off exercises that have been infrequently repeated due to the amount of effort required. Therefore, in this work, we developed a transparent and reproducible approach to update our data product while also enabling easier reproduction of future datasets. We distribute our 2015 product at the same resolution and formats as the prior product, and accompany it with a full set of replicable code and data for reconstruction. In this article we explain how the data was constructed, with links to the permanent DOIs where the data can be readily downloaded by the user community (Mehrabi et al. 2024; DOI: 10.5281/zenodo.11540554).