• Research

    Ongoing Projects

    Text Mining and Visualization Techniques to Study Online Communities
    In this project we are developing and evaluating various tools and techniques to automatically uncover social networks between online participants from just their digital footprints alone. Once discovered, these social networks can be used in a myriad of ways such as finding popular resources, sharing information within a network of trust, conducting viral marketing, identifying, analyzing consumers’ perceptions of products, and measuring the effectiveness of political campaigns in online and offline media.

    Context-Aware Information Retrieval
    This project investigates the potential of a tighter integration between searching for information on the Internet and using those results in academic writing. To test some of our design ideas, I have developed a web-based search system called Personal Information Research Assistant (PIRA) that can seamlessly and unobtrusively support information searching in the context of academic writing. One of the major benefits of such a system is that it allows the user to jump directly into the task of writing and to stay focused on his or her writing. As part of this project, we are using a rapid prototyping approach to explore the design space and to develop working prototypes that can provoke reactions, design suggestions and discussions about desirable functionalities and interfaces.

    • Collaborators: Michael B. Twidale (University of Illinois at Urbana-Champaign, USA)
    • Web prototype: http://writencite.com
    Mobility of User-Created Data across Social Networking Websites
    As Internet users are joining more online communities such as Twitter, Facebook and MySpace, they are also becoming more prolific content creators. But unfortunately, the task of managing and exchanging content between their various online social networking (SN) communities remains cumbersome and time consuming. Fortunately, through the use of web APIs (Application Program Interface) and web mashup tools users now have an easy method to integrate and exchange data across various websites. In this research, I am studying how the availability of these various web APIs and web mashup tools may influence users’ information behavior on various social networking websites.

    New Projects

    Large Scale Analysis of the Blogosphere
    Web blogs have been the focus of many recent Internet studies, however most of the studies have focused on identifying communities of bloggers who discuss similar topics. Our project seeks to identify and examine emerging online communities among the readers/commentators on a single blog. At present, there has been very little research in how and why a group of blog readers, who often post their comments anonymously, goes from being a group of strangers to a coherent group where members recognize and acknowledge each other and develop communal ties. The project will answer these and other related questions by using a combination of techniques from linguistic analysis, web mining, and social network analysis.
    Korean Internet Network Miner (KINM)
    This project seeks to develop an e-research tool called Korean Internet Network Miner (KINM) that can automatically discover online social networks from Korean language website. Once KINM is developed and tested, it will be used to analyze Korean blogging communities on the world-first citizen journalism site OhMyNews (http://www.ohmynews.com).

    • Collaborators: Han Woo PARK (Oxford Internet Institute, UK & YeungNam University, Korea) and Chung-Joo Chung (State University of New York at Buffalo)
    Just-in-time Information Retrieval for Bloggers
    The project attempts to develop a web-based system that will automatically generate links and show content from related blogs to a blogger as she/he writes her/his own entry. Our expectation is that this tool will inform the blogger in real time of other related discussions that are going on in the blogosphere and give her/him a chance to be part of the ‘global’ conversion in the blogosphere. We hypothesize that our tool will lead to more interactive and content-rich blog entries, in which bloggers are not just expressing their own opinions, but also addressing each other and building on each other’s arguments.

    • Student Collaborator: Justin Wong (University of Toronto, Canada)
Share |

Anatoliy Gruzd, PhD
Assistant Professor
School of Information Management
Faculty of Management
Dalhousie University
6100 University Avenue, Suite 4010
Halifax, Nova Scotia B3H 3J5
Canada

Phone: 902-494-6119
Fax: 902-494-2451
E-mail: gruzd [at] dal.ca
Homepage: http://AnatoliyGruzd.com
Twitter: http://twitter.com/dalprof
Social Media Lab:
http://SocialMediaLab.ca

Twitter Stream

Archive

ResearchBlogging.org