Bilong Shen

Ph.D. Student
Institute for Software Research
School of Computer Science
Carnegie Mellon University
Pittsburgh, PA 15213, USA

Email: shenbilong AT gmail dot com


I am currently a visiting scholar in Institute for Software Research, Carnegie Mellon University. My advisor is Prof. Kathleen M. Carley. I am a Ph.D student from Tsinghua University, China. My advisor is Prof. Weimin Zheng . My research interests mainly focus on Spatial time series Data Mining, Machine Learning, Deep Learning, and their applications in real world problem.


StepDeep: A Novel Spatial-temporal Mobility Event Prediction Framework based on Deep Neural Network.
Bilong Shen, Xiaodan Liang, Yufeng Ouyang, Weimin Zheng, Kathleen M. Carley
The 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2018), London, United Kingdom. 19 - 23 August 2018
V-Tree: Efficient kNN Search on Moving Objects with Road-Network Constraints
Bilong Shen, Ying Zhao, Guoliang Li , Weimin Zheng, Yue Qin, Bo Yuan, Yongming Rao
IEEE International Conference on Data Engineering (ICDE 2017), April 19-22, 2017, San Diego, California, USA
Urban Activity Mining Framework for Ride Sharing Systems based on Vehicular Social Networks
Bilong Shen, Weimin Zheng, Kathleen M. Carley
Networks and Spatial Economics, ISSN: 1566-113X (Print) 1572-9427 (Online)SCI 2.695 (Accepted, it will be published in 2019.)
Roo: Route Planning Algorithm for Ride Sharing Systems on Large-Scale Road Networks, BigComp 2019: 2019 IEEE International Conference on Big Data and Smart Computing, Kyoto, Japan, February 27-March 2, 2019
Bilong Shen, Bo Cao, Ying Zhao, Haojia Zuo, Weimin Zheng, Yan Huang
(Accepted, it will be published soon.)
OCEAN: Fast Discovery of High Utility Occupancy Itemsets
Bilong Shen, Zhaoduo Wen, Ying Zhao, Dongliang Zhou, Weimin Zheng
The Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD'16), April 19-22, 2016, Auckland, New Zealand
Dynamic Ridesharing
Bilong Shen, Yan Huang, Ying Zhao
Journal of SIGSPATIAL Special, Volume 7 Issue 3, November 2015, ACM New York, NY, USA
Survey on Dynamic Ridesharing in Big Data Era
Bilong Shen, Yan Huang, Ying Zhao, Weimin Zheng
Journal of Computer Research and Development
Abnormal Subspace Sparse PCA for Anomaly Detection and Interpretation
Xingyan Bin, Ying Zhao, Bilong Shen
in Proceedings of SIGKDD'15 Workshop on ODDx3. August 10-13, Hilton, Sydney


An Intelligent Position-identifying Device and Intelligent Position-identifying System, 201610037665.9, 105699974A
Bilong Shen, Sixing Wu


I build or contribute to a large variety of projects across data mining, frequent pattern mining, and deep learning application on spatial data, some of them are open sourced.
  • Spatial data mining technology on furture ride sharing system (Sep. 2013 to present)
    • a. Designed and implemented a smart framework for dynamic ride sharing system.
    • b. Devised an efficient and scalable tree index for moving objects on the road network, called V-Tree.
    • c. Proposed a novel method using V-Tree to compute kNN moving objects on road networks.
    This method is 1,000 to 100,000 faster than the current kNN search methods on road networks for moving objects on US road networks. This work was published in Proceedings of IEEE International Conference on Data Engineering (ICDE 2017). I was invited to the Innovation and Entrepreneurship Forum of China in 2016 (This work is the only Tsinghua University's student work that was represented in the forum).

  • Generate subsidy policy for Ride Sharing System by mining spatial-temporal data (Dec. 2014 to Jun. 2015)
    • a. Designed and led a team to implement a system for analyzing the hobby of drivers during the internship in 51yche Ride Sharing Company (The second largest ride sharing company in China).
    • b. Proposed a Ranking Method for passengers ranking.
    This work was applied in the ride sharing system of 51yche Ride Sharing Company. The analysis system can save more than 2 million dollors per year for the company.

  • Discover CoFlexi routes from Trajectories of Taxies (Dec. 2016 to Jun. 2017)
    • a. Proposed and implemented the clustering algorithm for generating the best routes for CoFlexi from trajectories of taxies.
    • b. Proposed a distance measure to solve the complex matching problem in ride sharing.
    The algorithm reduced the number of vehicles for the same number of requests by 21%.

  • Network Traffic Analysis for the Mobile Operator (Sep. 2013 to Sep. 2014)
    • a. Proposed a novel measure called "utility occupancy" for mobile operator data.
    • b. Derived an upper bound for utility occupancy.
    • c. Designed an efficient mining algorithm called OCEAN based on a fast implementation of utility list.
    This algorithm was used by Mobile Operator and published in the Proceedings of the Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)2016 (AR:39/307= 12.7).

Working Experience

  • Intern, 51yche Ride Sharing Company, 2015
  • Intern, Deloitte China, 2014