Current Projects

  1. Modeling Trust and Credibility in Microblogs

We develop credibility models to automate evaluation process of information credibility across different data sources such as social network, world-wide-web and semantic web space.  In the previous step of our research, we found that credibility perception of a person varies depending on one’s perspective in context or trust value.  Therefore, we try to model several user types as information sources or agents(disseminators) and define multiple contexts such as different topic spaces.  Currently, we are focusing on the credibility measurement in microblogs such as Twitter or LinkedIn.

  1. Discovering Patterns in Online Streams

Through web mining approach, we can discover sailent patterns in various forms such as peaks of communication in social media.  For instance, measuring keyword frequency distribution on Twitter mentions or updates can reveal a highlighted news event and, thus, we can further scrutinize the quantitative correlation between the online information stream and the real-world event.

[An example from]


  1. Interactive Data Visualization

Various data sets are mined by our automated crawling scripts and stored into relational(MySQL) or semantic(Jena SDB) databases.  The information collected (including metadata) are processed accordingly in order to visualize interactively to end users.  These visualization frameworks serve as smart interfaces that enable users efficiently find what they search for and help discover valuable information that they could not expect.  Data visualization frameworks are implemented accordingly in different environments (Processing, JavaScript, HTML5, etc.)

  1. Real Time Hand Pose Recognition

In this project, we develop an advanced method that predicts articulated hand poses in real-time with a single depth camera(Kinect), time-of-flight camera or other sensory devices.  We strive to achieve our approach in real-time in order to apply the technique to various Mixed (Augmented or Virtual) Reality applications as a novel interface.  [HCI, Machine Learning, Computer Vision, Computer Graphics]




  1. Marker-less Augmented Reality – Beyond PTAM.


 Previous Projects

  1. DynaBook (Test Project)

-FLEX based Dynamic Social-Network Visualization

-Interactive Graph Visualization for Facebook and Twitter

  1. SmartConnect: Optimizing Wireless Access to save Power [CS276 Project]
  • The ubiquity of Internet enabled smartphones and the availability of 3G and WiFi networks enables us to always remain connected. Applications on mobile phones such as Facebook, Gmail, provide us a convenient way of keeping in touch with our family and friends. RSS Readers and News Readers help us keep ourselves up to date with the latest news. But connecting the phone to the network consumes a lot of energy, forcing us to choose between connectivity and battery life. Our studies showed that applications connecting to the network periodically looking for updates, results in constant on-off transitions of the network system. Earlier studies [1, Bartender] have shown that switching the network system on and off is a very expensive process in terms of energy consumption. We observed that most of the applications running on the phone do not have realtime data transfer requirements. Hence we propose that requests from different applications must be buffered and sent synchronously. Doing so helps reduce the on-off transitions of the network system reducing the energy consumption dramatically. In order to verify our hypothesis, we implemented an android service which buffers requests from multiple network client applications.



  1. Need A Ride [CS290F Project]
  • Need-A-Ride is an Android based mobile social application in which people can ask a ride to their friends. It includes Google Map API, GPS functionality, Bluetooth sensor technique, Facebook graph API, and central database server to exchange additional data which are not provided by Facebook graph API.



  1. Exploring PBRT with various rendering effects [MAT594G Project – Fall 2011]
  • I explored a few basic techniques such as depth of field as well as other effects found from heuristic experiments using PBRT.


  1. Adaptive Interfaces for Marker-based Augmented Reality

[CS290I Project – Winter 2012]

project website : Go!
– Augmented reality(AR) techniques have been developed for more than a decade in a variety of fields including Computer Science and its possible application practices are being exponentially increased.  As augmented reality technology is being developed rapidly, we are not only facing its increased demand in the real world, but being asked for more advanced interfaces as well as smooth adaptation to merge the technology with real world.  In this project, we propose a few interfaces that can be utilized in both legacy platform and mobile environment.  We suggest four different user interfaces (recognition, reflection, sound and motion interactions) for experiencing more realistic augmented reality application.