Work Hard and Play Hard...

Jierui(Jerry) Xie

Ph.D Student
Department of Computer Science
Rensselaer Polytechnic Institute
110 8th Street
Troy, New York 12180
USA

Email:
----------------------------------------------------------------------

I am currently a Ph.D. student at Rensselaer Polytechnic Institute (RPI). My advisor is Boleslaw K. Szymanski.

In general, my research interests include data mining, machine learning and social network. I am interested in applying data mining and machine learning technologies to knowledge/pattern discovery, classification, clustering in Internet, sensor network and social network.

Since 2007, I've been involved in projects supported by International Technology Alliance (ITA). My work includes applying statistic models (HMM and BN) to recognizing human activities from sensor signals; Invariant feature-based classification; learning the distance function for categorical data classification; dimension reduction and embedding.

Recently I work on information diffusion and opinion consensus with naming game (a stochastic model) on large social network. I am interested in studying the evolution of on-line social network (e.g., blog, social tagging) through web mining and graph mining. I am also interested in applying these techniques to semantic web, e.g., ontology learning, matching, ranking, searching, evolution etc.

Recent Publication:

• J. Xie, M. Beigi. "A SCALE-INVARIANT LOCAL DESCRIPTOR FOR EVENT RECOGNITION IN 1D SENSOR SIGNALS", IEEE International Conference on Multimedia & Expo(ICME),Page(s):1226 - 1229, 2009
• J. Xie, et al., A Universal Utility Framework for Optimizing Sensor Network Performance, submit to IEEE PERCOM 2010
• J. Xie, et al., Supervised Dissimilarity Measures for Categorical Symbols(draft).

Internship:

• Jun.-Aug 2009, Summer Intern at IBM T.J. Watson Research Center and ARL, Mentor:
  David Wood. I work on a universal utility framework for optimizing sensor network
  performance. The framework could assist in formulation of a good utility function for
  solving various practical problems in sensor networks.
• Jun.-Aug 2008, Summer Intern at IBM TJ Watson Research Center, Hawthorne. I worked with
  Mandis Beigi on density based change detection and event recognition. We proposed a
  shape-base scale invariant feature descriptor, which takes advantage of scale space theory
  to detect multiple temporal scale events from various kinds of sensors (Infrared, seismic,
  accelerometer and acoustic). The work is accepted by ICME2009.

TA:

• Computer organization 2007
• Computation complexity 2008
• Machine learning 2009