Postdoctoral Fellow in Data Mining in Social Systems

I am looking for postdoctoral fellows the area of data management mining, preferably with an emphasis on social networks, social media, text mining, and recommender systems. Position is subject to the availability of funding. The fellow will be working in the Data Management and Mining Lab in the Department of Computer Science at the University of British Columbia, Vancouver.

The focal area of research for this position is exemplified by the following projects. Firstly, while tremendous advances have been made in the area of influence maximization in social networks with viral marketing as its main touted application, to date the research has seen a modest penetration in industry-strength viral marketing case studies. The aim of this project is to identify the limitations of known research advances in influence maximization that have prevented the techniques from being applied in practice, and devise techniques for taking viral marketing out of the lab.

A second project will focus on developing techniques for recommending novel and non-traditional kinds of “items” to “users” in unconventional settings. Is prediction accuracy (along with its add-ons such as diversity, serendipity etc) the only metric with which to gauge the efficacy of a recommender system? What should a company powering its business with a recommendation engine focus on in order to maximize revenue? What features should a product have to attract customers? What are some non-traditional settings where recommendations could benefit applications? These are just some of the questions this project addresses.

How can we detect fake claims and more generally fake news? What kind of intervention is effective in containing its propagation?

"Filter bubble" is a well-known problem in social media as well as in news consumption? Given a topic, can we automatically identify the stance taken by articles? What can we do to encourage a "balanced" consumption of news covering opposing viewpoints on a given issue?

A PhD in computer science or a related field is required. The successful candidate will have a strong publication record and the ability and drive to work independently as well as with a team of talented and passionate researchers. Interested applicants should contact Professor Laks V.S. Lakshmanan at [first-name AT department DOT university DOT ca]. See this for an idea of related publications. Salary will be commesurate with qualifications and experience. The fellow will benefit from the strong ties we enjoy with several industry collaborators such as AT&T Labs, Google, LinkedIn, Oracle, Technicolor, and Yahoo! Research, and Walmart Labs.