About me

I am currently a master student in the School of Intelligent Systems Engineering at Sun Yat-Sen University. My research interests are machine learning, mobility computing, and urban traffic data analytics. Recently, I focused on spatiotemporal (multivariate/multidimensional) time series imputation and prediction, and the related papers have been published in the top journal - Transportation Research Part C: Emerging Technologies.

Research Interests

  • Machine learning: Bayesian matrix and tensor factorization; variational auto-encoder (VAE); generative adversarial network (GAN).
  • Urban traffic data analytics: Missing data imputation; spatio-temporal time series prediction; anomaly detection; individual trip patterns discovery.

Selected Publications

  • Xinyu Chen, Zhaocheng He, Yixian Chen, Yuhuan Lu, Jiawei Wang (2019). Missing traffic data imputation and pattern discovery with a Bayesian augmented tensor factorization model. Transportation Research Part C: Emerging Technologies, 104: 66-77. [preprint] [doi] [slide] [data] [Matlab code]

  • Xinyu Chen, Zhaocheng He, Lijun Sun (2019). A Bayesian tensor decomposition approach for spatiotemporal traffic data imputation. Transportation Research Part C: Emerging Technologies, 98: 73-84. [preprint] [doi] [data] [Matlab code] [Imputation example in Jupyter notebook (Matlab)] [Jupyter notebook (Python)]

  • Xinyu Chen, Zhaocheng He, Jiawei Wang (2018). Spatial-temporal traffic speed patterns discovery and incomplete data recovery via SVD-combined tensor decomposition. Transportation Research Part C: Emerging Technologies, 86: 59-77. [doi] [data]

Open Sources

  • Urban traffic speed data set: This is our open traffic speed data set, which consists of 214 anonymous road segments within two months at 10-minute interval, and the speed observations were collected in Guangzhou, China.

  • transdim: This is our open project for transportation data imputation which covers the missing data imputation and spatiotemporal short-term (or long-term) traffic forecasting using a variety of machine learning and deep learning models. [25+ stars]

  • awesome-LaTeX-drawing: Drawing Bayesian networks, graphical models and framework for academic studies in LaTeX. [180+ stars]

  • academic-drawing: Providing source codes (including Matlab and Python) for presenting experiment results. [65+ stars]