Two SCS Alumni Receive ACM SIGKDD Dissertation Honors

Printer-friendly version

Amr Ahmed, a Yahoo! Research scientist who earned his PhD in Language Technologies from the Carnegie Mellon in 2011, is the winner of the prestigious 2012 Doctoral Dissertation Award from the Association for Computing Machinery's Special Interest Group on Knowledge Discovery and Data Mining (ACM SIGKDD). Lei Li, a post-doctoral researcher at the University of California, Berkeley, who earned a PhD in computer science in 2011, was the runner up for the award.

The ACM SIGKDD is the premier society for data mining and knowledge discovery research, and the ACM SIGKDD dissertation award is the highest distinction for a PhD in the area. Amr and Lei will be honored at the KDD Conference in Beijing, China, Aug. 12-16.

Amr, who also earned a M.Sc. in Machine Learning in 2009, will present a 15-minute summary of his dissertation entitled "Modeling Content and Users: Structured Probabilistic Representation and Scalable Online Inference Algorithms," during a special session at the conference.
    
Amr's thesis developed novel models for exploring and understanding the fast-evolving content of social media and scientific literature. "The goal of my research is to help users 'get the big picture' without drowning in a sea of irrelevant content," he said.

"It is a challenging goal to achieve especially in real-time, large-scale settings. For example, in social media my dissertation research can detect ideological bias in blog posts as well as analyze the temporal dynamics of events and stories in the news.

"Furthermore, we can show users articles spanning diverse ideological perspectives about any topic, e.g. conflicts in the Middle East, thus helping users to stay informed. As another example, in scientific literature the work we developed can discover papers that spawned new fields, track the evolution of those fields, and facilitate finding relevant papers using different modalities," Amr added.

"I believe the work Amr has done in his dissertation will have a significant impact on the general theory and algorithms of statistical modeling and large-scale data mining, and on practical web-scale information retrieval," Amr's adviser, Eric Xing, associate professor of machine learning, wrote in a letter nominating him for the dissertation award.

In his dissertation, "Fast Algorithms for Mining Co-evolving Time Series," Li developed novel algorithms for forecasting, clustering and missing-value imputation for time sequences in a broad spectrum of settings, from motion-capture sequences to data-center monitoring. Christos Faloutsos, professor of computer science, was Li's adviser.

Byron Spice | 412-268-9068 | bspice@cs.cmu.edu