» Go to news main

Feds boost high‑risk, high‑reward research at Dal with New Frontiers funding

Posted by Michele Charlton on June 7, 2021 in Research, Faculty, Awards, Big Data & Machine Learning
Clockwise from top left: Dr. Zhenyu Cheng, Dr. Paola Marignani, Dr. Evangelos Milios, Dr. Ingrid Waldron, Dr. Daniel Boyd. (Provided photos)
Clockwise from top left: Dr. Zhenyu Cheng, Dr. Paola Marignani, Dr. Evangelos Milios, Dr. Ingrid Waldron, Dr. Daniel Boyd. (Provided photos)

Five Dal researchers have received a $750,000 investment from the Government of Canada to advance their innovative ideas.

The funding comes from the New Frontiers in Research Fund (NFRF) Exploration competition, which has an objective of supporting high risk, high reward and interdisciplinary research. This year’s awards, with grants up to $250,000 over two years, are supporting 117 research projects across Canada with the potential to yield game-changing results in social, cultural, economic, health-related or technological areas.

“Congratulations to this talented group of researchers, whose bold ideas will have significant impact across Canada and around the globe,” says Alice Aiken, Dalhousie’s vice president research and innovation. “This investment is recognition of the outstanding quality of research at  Dalhousie, and helps us lead the way forward on solving some of the world’s most complex issues.”

One of the recipients is the Faculty of Computer Science's Dr. Evangelos Milios.

Researcher: Dr. Evangelos Milios
Co-PI: Dr. Evangelia Tastsoglou, Saint Mary’s University
Co-applicant: Dr. Eugena Kwon, Saint Mary’s University
 Computer Science
Visual analytics for text-intensive social science research on immigration

Text-intensive research in social sciences relies on the retrieval, organization, conceptualization and summarization of large amounts of text, with the aim of obtaining insights on social science research questions. Typically, social science researchers can only read and annotate limited amounts of text, so the amount of text data must be constrained by limiting the scope of the research question. In addition, retrieval of relevant text data is carried out by key term searches, which risks missing relevant documents using unanticipated vocabulary, and including irrelevant documents simply because they happen to include the search terms.

This project introduces a novel methodological paradigm in social science research that employs natural language processing and visual analytics to enable social scientists to retrieve and make sense of large document collections. From a computer science perspective, the transition from laboratory evaluations to addressing real-world problems and the close interdisciplinary collaboration will advance the state-of-the-art design of visual analytics (VA) systems aiming for usability by social scientists, enabling them to analyze much larger document sets than has been feasible to date.

Read the full artcile in Dal News.