A Context-Aware Topic Analysis Framework

Online social networks model and operate primarily on explicit relationships between
people. However, many high-value social activities are not mediated by relationships of
this kind. Instead, people who may not know each other nonetheless co-ordinate their
activities because they attend to similar cues in their cultural environment. This research
work aims at exploring ways to help social scenes and subcultures produce positive
cultural outcomes more reliably by examining how topic analysis techniques can be
applied to reveal and recommend opportunities for cultural participation. The goal of this
research work is to build an intelligent context-aware graph model that uses a mash-up of
information and semantic metadata from different sources about real entities. The
proposed framework aims at providing a new context-aware knowledge mining engine to
enhance the user experience by providing an optimal answer taking into account the four
main dimensions affecting the human judgments (i.e., Time, Place, Topic, and People).

Faculty Supervisor:

Ladan Tahvildari

Student:

Partner:

ODScore

Discipline:

Engineering

Sector:

Professional, scientific and technical services

University:

University of Waterloo

Program: