Current Research Area

  • Mining Social media data: This research area explores machine learning techniques such as graphical models and content analysis to explore approaches to effectively explore mining social data for understanding user level and aggregate trends in the data.
  • Mechanisms for aggregating social network data to make it accessible for Business use-cases
  • Understanding relationships and correlation between online data from varied sources
  • Social media monitoring techniques 

Previous Research

Research Projects

  • Joint Inference to Extract Soft Biometric Text Descriptors from Triage Images 
    Extracting various soft biometric features from Triage Images of Disaster Victims to form a Text Descriptor that describes the person in the image.
    Soft biometric features are related and combined using a Markov Network (Probabilistic graphical model) and Joint Inference is used for communication between features. 
    This work is closely related with the Lost Person Finder project at the National Library of Medicine, NIH, USA. 

  • LPFMobile: Prototype for Handheld device compatible version of Lost Person Finder 
    Designed and developed a mobile phone compatible version of the Lost Person Finder project as a Research Intern at NLM, NIH. 

  • BuddyCast Inc: Communication Protocol for Online Video Sharing
    Research and development of a prototype system for online video sharing software. The communication protocol architecture was setup and developed for a startup, BuddyCast Inc.

Publications

  • Niyati Chhaya, "Joint Inference for Extracting Text Descriptors from Triage Images of Mass Disaster Victims", AAAI-SIGART Doctoral Consortium, August 2011 Abstract
  • Niyati Chhaya and Tim Oates, "Robust Face Detection in Patient Triage Images", International Conference on Computer Vision Theory and Application (VISAPP,VISIGRPP), March 2011 PDF
  • Niyati Chhaya and Tim Oates, "Joint Inference of Soft Biometric Features", IEEE/ IAPR International Conference on Biometrics, March 2012

Presentations and Posters

  • Niyati Chhaya, "Joint Inference to Extract Soft Biometric Descriptions", Workshop for Women in Machine Learning, December 2011, to appear
  • Niyati Chhaya and Tim Oates, "Combinging Soft Biometrics from Images to Generate Text Descriptors", NIPS Workshop on Integrating Language and Vision, Poster and Spotlight talk, December 2011, to appear Abstract
  • Niyati Chhaya, "Joint Inference for Extracting Text Descriptors from Triage Images of Mass Disaster Victims", ACM Student Research Competition, Poster Session and Presentation 
    Grace Hopper Celebration of Women In Computing, 2011 Abstract
  • Niyati Chhaya and Tim Oates, "Joint Inference for Extracting Text Descriptors from Triage Images of Mass Disaster Victims", Mid-Atlantic Student Symposium on Speech, Language and Learning, 2011 Abstract
  • Niyati Chhaya, "Joint Inference for Extracting Text Descriptors from Images of Disaster Victims", Workshop for Women in Machine Learning, December 2010.

Scholarships and Awards

  • ACM Graduate Student Research Award, 3rd position, Grace Hopper Celebration of Women in Computing, 2011
  • NIPS Volunteer scholarship and Travel support (NIPS Workshop on Integerating Language and Vision), NIPS 2011
  • AAAI Doctoral Consortium and Student Scholarship 2011
  • ICML Student Volunteer Scholarship 2011
  • ACM Student Research Competition Participant Scholarship (Grace Hopper Celebration of Women in Computing) 2011
  • Travel Scholarship, Workshop for Women in Machine Learning, 2010, 2011
  • Research Observer and Volunteer, BHEPP Mass Disaster Drills, 2010,2009