Jump to Header Jump to Main Content Jump to Footer

Computer science graduate student pens award-winning thesis

Home > Blog > Computer science graduate student pens award-winning thesis

Computer science graduate student pens award-winning thesis

Ibrahim Almosallam, a computer science doctoral candidate, received MU's 2009 Distinguished Master's Thesis Award.

Ibrahim Almosallam, a computer science doctoral candidate in the College of Engineering is the recipient of the University of Missouri’s Distinguished Master’s Thesis Award.

According to George Justice, associate dean of the graduate school, Almosallam’s paper, “A New Adaptive Framework for Collaborative Filtering Prediction,” was chosen based on the clarity and power of his thesis abstract, as well as a letter of recommendation from his advisor — and the thesis itself.  Yi Shang, professor and the director of graduate studies in computer science serves as Almosallam’s advisor.

“I was shocked when I learned that I had received the award. It’s huge honor to receive this recognition from the University,” said Almosallam.

He conceived of his thesis in response to a challenge by the online feature film rental company Netflix, a competition aimed at improving the firm’s recommendation system.

There exists a trend by businesses to use computational intelligence and data mining methods to provide clientele with more personalized product or service recommendations. Known as collaborative filtering (CF), it has proved to be one of the most successful techniques for commercial service recommendations.

In his thesis, completed in Spring 2008 with Shang, Almosallam focused on improving the accuracy of CF in prediction and estimation applications for a large-scale real-world problem: the million-dollar Netflix Challenge.

When he wrote the paper about his performance in the ongoing Challenge, Almosallam’s system with its unique adaptive framework and methods, was capable of outperforming the best existing CF. In his last post at the end of 2007 he had achieving a 5.6% improvement over Netflix’s Cinematch system. That score placed him in the top 0.5% of 32,000 teams in this worldwide competition.

Almosallam has moved on to other projects within the department, including wireless sensor localization and protein structure prediction.

Back to Top

Enter your keyword