Zefang Shen

I develop AI-based solutions and conduct novel Machine Learning research. See my talk at the Australian Academy of Science (below) for a glimpse into my work.

I am experienced in building data science pipelines using Python and R. I applied various machine learning algorithms (support vector machines, random forests, XGBoost, Gaussian Processes) and deep learning architectures (convolutional neural networks, LSTMs, attention-based models) to deliver high-impact solutions. My work has resulted in high-quality publications and real-world applications.

My research tackles challenging problems in machine learning for spectroscopy and robotics, resulting in publications in top tier journals such as ISPRS Journal of Photogrammetry and Remote Sensing (IF12.7, 2022), Earth-Science Reviews (IF12.1), Global Change Biology (IF11.6), Communications earth & environment (IF7.9), SOIL(IF6.8), etc. Of particular distinction is my receipt of the Journal of Mechanical Design 2018 Editors' Choice Paper Award given to one article among all in 2018 for fundamental contribution in robotic design.

I am passionate about transforming research outputs into real-world solutions. With my full-stack data science skills (backend: Python Django; frontend: Bootstrap/React; Deployment: AWS), I implemented machine learning pipelines for the modeling of spectral data into a web application, which resulted in the world’s first global service for estimating of soil organic carbon with spectra.

Highlights

Artificial Intelligence for Soil Science

Australian Academy of Science, Canberra, Australia. 09/04/2024

I was honored to be invited by the Australian Academy of Science to share my expertise in developing AI for Soil Science at the public speaker series: The Journey of Australian Science. The talk focused on challenges in soil spectroscopic modelling and our solutions. I also shared my thoughts on future directions of AI research and applications in Soil Science.

Learn more about the event here.

ASME Journal of Mechanical Design 2018 Editors' Choice Paper Award

IDETC/CIE, California, USA. 08/2019

This award is given to only one paper each year from all articles published in the prestigious Journal of Mechanical Design. The selection of this paper was based on the recommendations of the Associate and Guest Editors and guided by the following criteria (i) fundamental value of the contribution, (ii) expectation of archival value (e.g., expected number of citations), (iii) practical relevance to mechanical design, and (iv) quality of presentation. Please find the awarded paper here.

News

Jul 24, 2024	New paper published in Science of the Total Environment. The research tested MIR spectra for characterising carbon content in coastal marine soils.
Jul 01, 2024	New paper in Earth-Science Reviews using Transfer Learning to address the challenging localisation problem in soil spectroscopy modelling is now online.
Apr 10, 2024	The Journey of Australian Science: presented AI solutions for Soil Science at the Australian Academy of Science.
Dec 02, 2023	New paper in Global Change Biology studying the Carbon sequestration potential of Australian soils is now online.
Jul 05, 2023	Webinar: Artificial intelligence and machine learning in soil spectroscopic modelling. Shared my experience building AI-based solutions for soil spectroscopic modelling in this webinar organised by the Global Soil Partnership, Food and Agriculture Organisation of the United Nations.

Selected publications

Earth-Sicence Rev.

An imperative for soil spectroscopic modelling is to think global but fit local with transfer learning

Raphael A Viscarra Rossel, Zefang Shen, Leonardo Ramirez Lopez, and 8 more authors

Earth-Science Reviews, 2024

Bib HTML

@article{rossel2024imperative,
  title = {An imperative for soil spectroscopic modelling is to think global but fit local with transfer learning},
  author = {Rossel, Raphael A Viscarra and Shen, Zefang and Lopez, Leonardo Ramirez and Behrens, Thorsten and Shi, Zhou and Wetterlind, Johanna and Sudduth, Kenneth A and Stenberg, Bo and Guerrero, Cesar and Gholizadeh, Asa and others},
  journal = {Earth-Science Reviews},
  pages = {104797},
  year = {2024},
  publisher = {Elsevier},
  url = {https://doi.org/10.1016/j.earscirev.2024.104797},
}

SOIL

Miniaturised visible and near-infrared spectrometers for assessing soil health indicators in mine site rehabilitation

Zefang Shen, Haylee D’Agui, Lewis Walden, and 8 more authors

Soil, 2022

Bib HTML

@article{shen2022miniaturised,
  title = {Miniaturised visible and near-infrared spectrometers for assessing soil health indicators in mine site rehabilitation},
  author = {Shen, Zefang and D'Agui, Haylee and Walden, Lewis and Zhang, Mingxi and Yiu, Tsoek Man and Dixon, Kingsley and Nevill, Paul and Cross, Adam and Matangulu, Mohana and Hu, Yang and others},
  journal = {Soil},
  volume = {8},
  number = {2},
  pages = {467--486},
  year = {2022},
  publisher = {Copernicus Publications G{\"o}ttingen, Germany},
}

P&RS

Deep transfer learning of global spectra for local soil carbon monitoring

Zefang Shen, Leonardo Ramirez-Lopez, Thorsten Behrens, and 8 more authors

ISPRS Journal of Photogrammetry and Remote Sensing, 2022

Bib HTML

@article{shen2022deep,
  title = {Deep transfer learning of global spectra for local soil carbon monitoring},
  author = {Shen, Zefang and Ramirez-Lopez, Leonardo and Behrens, Thorsten and Cui, Lei and Zhang, Mingxi and Walden, Lewis and Wetterlind, Johanna and Shi, Zhou and Sudduth, Kenneth A and Baumann, Philipp and others},
  journal = {ISPRS Journal of Photogrammetry and Remote Sensing},
  volume = {188},
  pages = {190--200},
  year = {2022},
  publisher = {Elsevier},
}

Sci. Rep.

Automated spectroscopic modelling with optimised convolutional neural networks

Zefang Shen, and RA Viscarra Rossel

Scientific Reports, 2021

Bib HTML

@article{shen2021automated,
  title = {Automated spectroscopic modelling with optimised convolutional neural networks},
  author = {Shen, Zefang and Viscarra Rossel, RA},
  journal = {Scientific Reports},
  volume = {11},
  number = {1},
  pages = {208},
  year = {2021},
  publisher = {Nature Publishing Group UK London},
}

ASME JMD

An integrated type and dimensional synthesis method to design one degree-of-freedom planar linkages with only revolute joints for exoskeletons

Zefang Shen, Garry Allison, and Lei Cui

Journal of Mechanical Design, 2018

Bib HTML

@article{shen2018integrated,
  title = {An integrated type and dimensional synthesis method to design one degree-of-freedom planar linkages with only revolute joints for exoskeletons},
  author = {Shen, Zefang and Allison, Garry and Cui, Lei},
  journal = {Journal of Mechanical Design},
  volume = {140},
  number = {9},
  pages = {092302},
  year = {2018},
  publisher = {American Society of Mechanical Engineers},
}