Gunhee Kim, who completed his Ph.D. in computer science last year and is now a post-doctoral researcher at Disney Research Pittsburgh, has received the 2014 SIGKDD Dissertation Award. The award, presented by the Association for Computing Machinery's Special Interest Group on Knowledge Discovery and Data Mining, recognizes outstanding work by graduate students in the areas of data science, machine learning and data mining.
Mladen Kolar, who earned his Ph.D. in machine learning and is now an assistant professor of econometrics and statistics at the University of Chicago, received an honorable mention. Both Kim and Kolar were advised by Eric Xing, professor of machine learning.
Kim’s thesis, “Reconstruction and Applications of Web Photo Storylines,” addressed the ever-growing online volume of user-generated images and videos. He developed scalable machine learning and optimization techniques to understand the hidden stories underlying this visual data, including modeling of temporal trends in image collections, extracting story structures from large visual data and performing visual brand and market analyses. It represents the first large-scale data-mining of image and video data; previous work has been largely on text data, rather than the modality that people are most familiar with --- what they see with their eyes.