Skills

  • Experienced with Python, R, SQL. Familiar with Java, Objective-C
  • Machine Learning Algorithm: Support Vector Machine, Decision Trees, Online Learning, Feature Engineering, NLP, etc. 
  • Packages/Libraries:Python Scikit-Learning, Tensorflow, Weka Packages, R Machine Learning Package
  • Data Analysis Techniques: Mixed Model, PCA, ANOVA, Regression, etc.
  • Smartphone Sensors: Accelerometer, Barometer, Magnetometer, Gyroscope, etc.
  • Operating System: Linux/Unix

Education

2011 - 2017 


2007 - 2010


2003 - 2007
Graduate Center, The City University of New York, US
Ph.D in Computer Science
Advisor:  Dr. Hanghang Tong

Zhejiang University, China
MS. in Computer Science
Member of the Lab of Very Large Information System

Shanghai Jiaotong University, China
BS. in Computer Science

Projects

  • Data Mining with Smartphone Sensors     (2013 - 2017)   

    • Accelerometer-based Activity Recognition with Smartphones 
      In this project  we utilize smartphone sensor: accelerometer to learn the activity patterns of users'. The activities we try to detect are 6 indoor activities: walking, jogging, walking upstairs, walking downstairs, sitting, taking elevator, standing. The system takes smartphone accelerometer readings as input and output is prediction of current activity. Decision Trees and Hidden Markov Chain are the main machine learning method we use. We also implement the iOS App. The work is demonstrated in the Conference on Information and Knowledge Management in 2014.

    • Travel Mode Detection with Smartphone Sensors 
      In this project we develop methods to detect travel mode in urban transportation with the data from multimodal sensors on smartphones. Instead of using the traditional online sensor data such as GPS and GRS data, we learn the patterns from motion sensors and environment sensors on smartphones. The system is with hierarchical framework: first layer classifier to detect wheeled travel modes (Train, Car, Bus, Bike) and unwheeled travel modes (Walking, Jogging), second layer classifier to detect the exact travel mode among the 6 categories. The work is demonstrated on the Transportation Research Board Annual Meeting in Washington in 2015. The system also employs online learning method and segmentation techniques so that the learned model is adaptive to current users, the system has quick response, and the smartphone app is energy-efficiency. This part of work is to be published. This project is cooperated with Dr. Qing He and Hernan Caceres from University at Buffalo in this project. 

  • Stochastic Modeling in Cognitive Radio Network  (2012 - 2013)            

    In this project, I work with Suman Bhunia, Dr. Shamik Sengupta and Dr. Felisa Vázquez-Abad on preventing jamming attacks with honeypot in  cognitive radio network. We develop queuing model for CRNs, which pose unique modeling challenges due to their periodic sensing and transmission cycles.  We simulated the CRNs under attack from jammers and introduce a series of strategies for honey-node assignment to combat these attacks, and assess the performance of each strategy. 

  • Multi-Choice Knapsack Problem in Social Swarming Application (2011 - 2012):

    The Project aims to find the most efficient yet economic way to choose resources for social swarming. Participants equipped with 3G and Wifi-capable are tasked to provide reports (possibly voluminous ones that include full-motion video) about their immediate environment to a central coordinator.  I work with Dr. Amotz Bar-Noy and Dr. Peter Terlecky. We modeled the scenario with multi-choice knapsack problem and explored solutions such as dynamic programming, greedy, semi-greedy, and heuristic methods such as Tabu search. 

Publications

  • Nancy Y. Kiang, Robert M. Haralick, Art Diky, Xing Su, Benjamin I. Cook, Climate-vegetation classification: Linear Manifold Clustering as a means objectively to identify physical bifurcations, climate change, and GCM climate biases. NASA Carbon System and Ecosystem Workshop, 2015. [Abstract]
  • Xing Su, Hanghang Tong, and Ping Ji. "Accelerometer-based Activity Recognition on Smartphone." Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management. ACM, 2014. [PDF]
  • Xing Su, Hanghang Tong, and Ping Ji. "Activity recognition with smartphone sensors." Tsinghua Science and Technology 19.3 (2014): 235-249. [PDF]
  • Bhunia, Suman, Xing Su, Shamik Sengupta, and Felisa Vázquez-Abad. "Stochastic Model for Cognitive Radio Networks under Jamming Attacks and Honeypot-Based Prevention." In Distributed Computing and Networking, pp. 438-452. Springer Berlin Heidelberg, 2014. [PDF]
  • Xing Su, Xiaohu Yang, Juefeng Li and Di Wu. "Parallel iterative reengineering model of legacy systems."Systems, Man and Cybernetics, 2009. SMC 2009. IEEE International Conference on. IEEE, 2009. [PDF]